hohn/codeql
1
0
Fork 0
mirror of https://github.com/github/codeql.git synced 2025-12-17 01:03:14 +01:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #19984 from jketema/jketema/sec-shared

Browse Source 
Make a proper shared library out of the concept related libraries
This commit is contained in:
Jeroen Ketema
2025-07-17 13:25:33 +02:00
committed by GitHub
parent 53e1939b60 1990438376
commit acc66c7b58
66 changed files with 530 additions and 2388 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
config/identical-files.json
View File

@@ -231,35 +231,10 @@
"java/ql/src/experimental/Security/CWE/CWE-400/LocalThreadResourceAbuse.qhelp", "java/ql/src/experimental/Security/CWE/CWE-400/LocalThreadResourceAbuse.qhelp",
"java/ql/src/experimental/Security/CWE/CWE-400/ThreadResourceAbuse.qhelp" "java/ql/src/experimental/Security/CWE/CWE-400/ThreadResourceAbuse.qhelp"
], ],
"CryptoAlgorithms Python/JS/Ruby": [
"javascript/ql/lib/semmle/javascript/security/CryptoAlgorithms.qll",
"python/ql/lib/semmle/python/concepts/CryptoAlgorithms.qll",
"ruby/ql/lib/codeql/ruby/security/CryptoAlgorithms.qll",
"rust/ql/lib/codeql/rust/security/CryptoAlgorithms.qll"
],
"CryptoAlgorithmNames Python/JS/Ruby": [
"javascript/ql/lib/semmle/javascript/security/internal/CryptoAlgorithmNames.qll",
"python/ql/lib/semmle/python/concepts/internal/CryptoAlgorithmNames.qll",
"ruby/ql/lib/codeql/ruby/security/internal/CryptoAlgorithmNames.qll",
"rust/ql/lib/codeql/rust/security/internal/CryptoAlgorithmNames.qll"
],
"SensitiveDataHeuristics Python/JS": [
"javascript/ql/lib/semmle/javascript/security/internal/SensitiveDataHeuristics.qll",
"python/ql/lib/semmle/python/security/internal/SensitiveDataHeuristics.qll",
"ruby/ql/lib/codeql/ruby/security/internal/SensitiveDataHeuristics.qll",
"swift/ql/lib/codeql/swift/security/internal/SensitiveDataHeuristics.qll",
"rust/ql/lib/codeql/rust/security/internal/SensitiveDataHeuristics.qll"
],
"IncompleteUrlSubstringSanitization": [ "IncompleteUrlSubstringSanitization": [
"javascript/ql/src/Security/CWE-020/IncompleteUrlSubstringSanitization.qll", "javascript/ql/src/Security/CWE-020/IncompleteUrlSubstringSanitization.qll",
"ruby/ql/src/queries/security/cwe-020/IncompleteUrlSubstringSanitization.qll" "ruby/ql/src/queries/security/cwe-020/IncompleteUrlSubstringSanitization.qll"
], ],
"Concepts Python/Ruby/JS": [
"python/ql/lib/semmle/python/internal/ConceptsShared.qll",
"ruby/ql/lib/codeql/ruby/internal/ConceptsShared.qll",
"javascript/ql/lib/semmle/javascript/internal/ConceptsShared.qll",
"rust/ql/lib/codeql/rust/internal/ConceptsShared.qll"
],
"ApiGraphModels": [ "ApiGraphModels": [
"javascript/ql/lib/semmle/javascript/frameworks/data/internal/ApiGraphModels.qll", "javascript/ql/lib/semmle/javascript/frameworks/data/internal/ApiGraphModels.qll",
"ruby/ql/lib/codeql/ruby/frameworks/data/internal/ApiGraphModels.qll", "ruby/ql/lib/codeql/ruby/frameworks/data/internal/ApiGraphModels.qll",

1
javascript/ql/lib/qlpack.yml
View File

@@ -6,6 +6,7 @@ extractor: javascript
library: true library: true
upgrades: upgrades upgrades: upgrades
dependencies: dependencies:
codeql/concepts: ${workspace}
codeql/dataflow: ${workspace} codeql/dataflow: ${workspace}
codeql/mad: ${workspace} codeql/mad: ${workspace}
codeql/regex: ${workspace} codeql/regex: ${workspace}

6
javascript/ql/lib/semmle/javascript/Concepts.qll
View File

@@ -5,7 +5,11 @@
*/ */
import javascript import javascript
private import semmle.javascript.dataflow.internal.sharedlib.DataFlowArg
private import codeql.threatmodels.ThreatModels private import codeql.threatmodels.ThreatModels
private import codeql.concepts.ConceptsShared
private module ConceptsShared = ConceptsMake<Location, JSDataFlow>;
/** /**
* A data flow source, for a specific threat-model. * A data flow source, for a specific threat-model.
@@ -206,7 +210,7 @@ abstract class PersistentWriteAccess extends DataFlow::Node {
* Provides models for cryptographic things. * Provides models for cryptographic things.
*/ */
module Cryptography { module Cryptography {
private import semmle.javascript.internal.ConceptsShared::Cryptography as SC private import ConceptsShared::Cryptography as SC
/** /**
* A data-flow node that is an application of a cryptographic algorithm. For example, * A data-flow node that is an application of a cryptographic algorithm. For example,

2
javascript/ql/lib/semmle/javascript/frameworks/CryptoLibraries.qll
View File

@@ -4,7 +4,7 @@
import javascript import javascript
import semmle.javascript.Concepts::Cryptography import semmle.javascript.Concepts::Cryptography
private import semmle.javascript.security.internal.CryptoAlgorithmNames private import codeql.concepts.internal.CryptoAlgorithmNames
/** /**
* A key used in a cryptographic algorithm. * A key used in a cryptographic algorithm.

7
javascript/ql/lib/semmle/javascript/internal/ConceptsImports.qll
View File

@@ -1,7 +0,0 @@
/**
* This file contains imports required for the JavaScript version of `ConceptsShared.qll`.
* Since they are language-specific, they can't be placed directly in that file, as it is shared between languages.
*/
import semmle.javascript.dataflow.DataFlow::DataFlow as DataFlow
import semmle.javascript.security.CryptoAlgorithms as CryptoAlgorithms

181
javascript/ql/lib/semmle/javascript/internal/ConceptsShared.qll
View File

@@ -1,181 +0,0 @@
/**
* Provides Concepts which are shared across languages.
*
* Each language has a language specific `Concepts.qll` file that can import the
* shared concepts from this file. A language can either re-export the concept directly,
* or can add additional member-predicates that are needed for that language.
*
* Moving forward, `Concepts.qll` will be the staging ground for brand new concepts from
* each language, but we will maintain a discipline of moving those concepts to
* `ConceptsShared.qll` ASAP.
*/
private import ConceptsImports
/**
* Provides models for cryptographic concepts.
*
* Note: The `CryptographicAlgorithm` class currently doesn't take weak keys into
* consideration for the `isWeak` member predicate. So RSA is always considered
* secure, although using a low number of bits will actually make it insecure. We plan
* to improve our libraries in the future to more precisely capture this aspect.
*/
module Cryptography {
class CryptographicAlgorithm = CryptoAlgorithms::CryptographicAlgorithm;
class EncryptionAlgorithm = CryptoAlgorithms::EncryptionAlgorithm;
class HashingAlgorithm = CryptoAlgorithms::HashingAlgorithm;
class PasswordHashingAlgorithm = CryptoAlgorithms::PasswordHashingAlgorithm;
/**
* A data flow node that is an application of a cryptographic algorithm. For example,
* encryption, decryption, signature-validation.
*
* Extend this class to refine existing API models. If you want to model new APIs,
* extend `CryptographicOperation::Range` instead.
*/
class CryptographicOperation extends DataFlow::Node instanceof CryptographicOperation::Range {
/** Gets the algorithm used, if it matches a known `CryptographicAlgorithm`. */
CryptographicAlgorithm getAlgorithm() { result = super.getAlgorithm() }
/** Gets the data flow node where the cryptographic algorithm used in this operation is configured. */
DataFlow::Node getInitialization() { result = super.getInitialization() }
/** Gets an input the algorithm is used on, for example the plain text input to be encrypted. */
DataFlow::Node getAnInput() { result = super.getAnInput() }
/**
* Gets the block mode used to perform this cryptographic operation.
*
* This predicate is only expected to have a result if two conditions hold:
* 1. The operation is an encryption operation, i.e. the algorithm used is an `EncryptionAlgorithm`, and
* 2. The algorithm used is a block cipher (not a stream cipher).
*
* If either of these conditions do not hold, then this predicate should have no result.
*/
BlockMode getBlockMode() { result = super.getBlockMode() }
}
/** Provides classes for modeling new applications of a cryptographic algorithms. */
module CryptographicOperation {
/**
* A data flow node that is an application of a cryptographic algorithm. For example,
* encryption, decryption, signature-validation.
*
* Extend this class to model new APIs. If you want to refine existing API models,
* extend `CryptographicOperation` instead.
*/
abstract class Range extends DataFlow::Node {
/** Gets the data flow node where the cryptographic algorithm used in this operation is configured. */
abstract DataFlow::Node getInitialization();
/** Gets the algorithm used, if it matches a known `CryptographicAlgorithm`. */
abstract CryptographicAlgorithm getAlgorithm();
/** Gets an input the algorithm is used on, for example the plain text input to be encrypted. */
abstract DataFlow::Node getAnInput();
/**
* Gets the block mode used to perform this cryptographic operation.
*
* This predicate is only expected to have a result if two conditions hold:
* 1. The operation is an encryption operation, i.e. the algorithm used is an `EncryptionAlgorithm`, and
* 2. The algorithm used is a block cipher (not a stream cipher).
*
* If either of these conditions do not hold, then this predicate should have no result.
*/
abstract BlockMode getBlockMode();
}
}
/**
* A cryptographic block cipher mode of operation. This can be used to encrypt
* data of arbitrary length using a block encryption algorithm.
*/
class BlockMode extends string {
BlockMode() {
this =
[
"ECB", "CBC", "GCM", "CCM", "CFB", "OFB", "CTR", "OPENPGP",
"XTS", // https://csrc.nist.gov/publications/detail/sp/800-38e/final
"EAX" // https://en.wikipedia.org/wiki/EAX_mode
]
}
/** Holds if this block mode is considered to be insecure. */
predicate isWeak() { this = "ECB" }
/** Holds if the given string appears to match this block mode. */
bindingset[s]
predicate matchesString(string s) { s.toUpperCase().matches("%" + this + "%") }
}
}
/** Provides classes for modeling HTTP-related APIs. */
module Http {
/** Provides classes for modeling HTTP clients. */
module Client {
/**
* A data flow node that makes an outgoing HTTP request.
*
* Extend this class to refine existing API models. If you want to model new APIs,
* extend `Http::Client::Request::Range` instead.
*/
class Request extends DataFlow::Node instanceof Request::Range {
/**
* Gets a data flow node that contributes to the URL of the request.
* Depending on the framework, a request may have multiple nodes which contribute to the URL.
*/
DataFlow::Node getAUrlPart() { result = super.getAUrlPart() }
/** Gets a string that identifies the framework used for this request. */
string getFramework() { result = super.getFramework() }
/**
* Holds if this request is made using a mode that disables SSL/TLS
* certificate validation, where `disablingNode` represents the point at
* which the validation was disabled, and `argumentOrigin` represents the origin
* of the argument that disabled the validation (which could be the same node as
* `disablingNode`).
*/
predicate disablesCertificateValidation(
DataFlow::Node disablingNode, DataFlow::Node argumentOrigin
) {
super.disablesCertificateValidation(disablingNode, argumentOrigin)
}
}
/** Provides a class for modeling new HTTP requests. */
module Request {
/**
* A data flow node that makes an outgoing HTTP request.
*
* Extend this class to model new APIs. If you want to refine existing API models,
* extend `Http::Client::Request` instead.
*/
abstract class Range extends DataFlow::Node {
/**
* Gets a data flow node that contributes to the URL of the request.
* Depending on the framework, a request may have multiple nodes which contribute to the URL.
*/
abstract DataFlow::Node getAUrlPart();
/** Gets a string that identifies the framework used for this request. */
abstract string getFramework();
/**
* Holds if this request is made using a mode that disables SSL/TLS
* certificate validation, where `disablingNode` represents the point at
* which the validation was disabled, and `argumentOrigin` represents the origin
* of the argument that disabled the validation (which could be the same node as
* `disablingNode`).
*/
abstract predicate disablesCertificateValidation(
DataFlow::Node disablingNode, DataFlow::Node argumentOrigin
);
}
}
}
}

114
javascript/ql/lib/semmle/javascript/security/CryptoAlgorithms.qll
View File

@@ -1,117 +1,5 @@
/** /**
* Provides classes modeling cryptographic algorithms, separated into strong and weak variants. * Provides classes modeling cryptographic algorithms, separated into strong and weak variants.
*
* The classification into strong and weak are based on Wikipedia, OWASP and Google (2021).
*/ */
private import internal.CryptoAlgorithmNames import codeql.concepts.CryptoAlgorithms
/**
* A cryptographic algorithm.
*/
private newtype TCryptographicAlgorithm =
MkHashingAlgorithm(string name, boolean isWeak) {
isStrongHashingAlgorithm(name) and isWeak = false
or
isWeakHashingAlgorithm(name) and isWeak = true
} or
MkEncryptionAlgorithm(string name, boolean isWeak) {
isStrongEncryptionAlgorithm(name) and isWeak = false
or
isWeakEncryptionAlgorithm(name) and isWeak = true
} or
MkPasswordHashingAlgorithm(string name, boolean isWeak) {
isStrongPasswordHashingAlgorithm(name) and isWeak = false
or
isWeakPasswordHashingAlgorithm(name) and isWeak = true
}
/**
* Gets the most specific `CryptographicAlgorithm` that matches the given `name`.
* A matching algorithm is one where the name of the algorithm matches the start of name, with allowances made for different name formats.
* In the case that multiple `CryptographicAlgorithm`s match the given `name`, the algorithm(s) with the longest name will be selected. This is intended to select more specific versions of algorithms when multiple versions could match - for example "SHA3_224" matches against both "SHA3" and "SHA3224", but the latter is a more precise match.
*/
bindingset[name]
private CryptographicAlgorithm getBestAlgorithmForName(string name) {
result =
max(CryptographicAlgorithm algorithm |
algorithm.getName() =
[
name.toUpperCase(), // the full name
name.toUpperCase().regexpCapture("^([\\w]+)(?:-.*)?$", 1), // the name prior to any dashes or spaces
name.toUpperCase().regexpCapture("^([A-Z0-9]+)(?:(-|_).*)?$", 1) // the name prior to any dashes, spaces, or underscores
].regexpReplaceAll("[-_ ]", "") // strip dashes, underscores, and spaces
|
algorithm order by algorithm.getName().length()
)
}
/**
* A cryptographic algorithm.
*/
abstract class CryptographicAlgorithm extends TCryptographicAlgorithm {
/** Gets a textual representation of this element. */
string toString() { result = this.getName() }
/**
* Gets the normalized name of this algorithm (upper-case, no spaces, dashes or underscores).
*/
abstract string getName();
/**
* Holds if the name of this algorithm is the most specific match for `name`.
* This predicate matches quite liberally to account for different ways of formatting algorithm names, e.g. using dashes, underscores, or spaces as separators, including or not including block modes of operation, etc.
*/
bindingset[name]
predicate matchesName(string name) { this = getBestAlgorithmForName(name) }
/**
* Holds if this algorithm is weak.
*/
abstract predicate isWeak();
}
/**
* A hashing algorithm such as `MD5` or `SHA512`.
*/
class HashingAlgorithm extends MkHashingAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
HashingAlgorithm() { this = MkHashingAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
}
/**
* An encryption algorithm such as `DES` or `AES512`.
*/
class EncryptionAlgorithm extends MkEncryptionAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
EncryptionAlgorithm() { this = MkEncryptionAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
/** Holds if this algorithm is a stream cipher. */
predicate isStreamCipher() { isStreamCipher(name) }
}
/**
* A password hashing algorithm such as `PBKDF2` or `SCRYPT`.
*/
class PasswordHashingAlgorithm extends MkPasswordHashingAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
PasswordHashingAlgorithm() { this = MkPasswordHashingAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
}

2
javascript/ql/lib/semmle/javascript/security/SensitiveActions.qll
View File

@@ -10,7 +10,7 @@
*/ */
import javascript import javascript
import semmle.javascript.security.internal.SensitiveDataHeuristics import codeql.concepts.internal.SensitiveDataHeuristics
private import HeuristicNames private import HeuristicNames
/** An expression that might contain sensitive data. */ /** An expression that might contain sensitive data. */

84
javascript/ql/lib/semmle/javascript/security/internal/CryptoAlgorithmNames.qll
View File

@@ -1,84 +0,0 @@
/**
* Names of cryptographic algorithms, separated into strong and weak variants.
*
* The names are normalized: upper-case, no spaces, dashes or underscores.
*
* The names are inspired by the names used in real world crypto libraries.
*
* The classification into strong and weak are based on Wikipedia, OWASP and Google (2021).
*/
/**
* Holds if `name` corresponds to a strong hashing algorithm.
*/
predicate isStrongHashingAlgorithm(string name) {
name =
[
// see https://cryptography.io/en/latest/hazmat/primitives/cryptographic-hashes/#blake2
// and https://www.blake2.net/
"BLAKE2", "BLAKE2B", "BLAKE2S",
// see https://github.com/BLAKE3-team/BLAKE3
"BLAKE3",
//
"DSA", "ED25519", "ES256", "ECDSA256", "ES384", "ECDSA384", "ES512", "ECDSA512", "SHA2",
"SHA224", "SHA256", "SHA384", "SHA512", "SHA3", "SHA3224", "SHA3256", "SHA3384", "SHA3512",
// see https://cryptography.io/en/latest/hazmat/primitives/cryptographic-hashes/#cryptography.hazmat.primitives.hashes.SHAKE128
"SHAKE128", "SHAKE256",
// see https://cryptography.io/en/latest/hazmat/primitives/cryptographic-hashes/#sm3
"SM3",
// see https://security.stackexchange.com/a/216297
"WHIRLPOOL",
]
}
/**
* Holds if `name` corresponds to a weak hashing algorithm.
*/
predicate isWeakHashingAlgorithm(string name) {
name =
[
"HAVEL128", "MD2", "MD4", "MD5", "PANAMA", "RIPEMD", "RIPEMD128", "RIPEMD256", "RIPEMD160",
"RIPEMD320", "SHA0", "SHA1"
]
}
/**
* Holds if `name` corresponds to a strong encryption algorithm.
*/
predicate isStrongEncryptionAlgorithm(string name) {
name =
[
"AES", "AES128", "AES192", "AES256", "AES512", "AES-128", "AES-192", "AES-256", "AES-512",
"ARIA", "BLOWFISH", "BF", "ECIES", "CAST", "CAST5", "CAMELLIA", "CAMELLIA128", "CAMELLIA192",
"CAMELLIA256", "CAMELLIA-128", "CAMELLIA-192", "CAMELLIA-256", "CHACHA", "GOST", "GOST89",
"IDEA", "RABBIT", "RSA", "SEED", "SM4"
]
}
/**
* Holds if `name` corresponds to a weak encryption algorithm.
*/
predicate isWeakEncryptionAlgorithm(string name) {
name =
[
"DES", "3DES", "DES3", "TRIPLEDES", "DESX", "TDEA", "TRIPLEDEA", "ARC2", "RC2", "ARC4", "RC4",
"ARCFOUR", "ARC5", "RC5"
]
}
/**
* Holds if `name` corresponds to a strong password hashing algorithm.
*/
predicate isStrongPasswordHashingAlgorithm(string name) {
name = ["ARGON2", "PBKDF2", "BCRYPT", "SCRYPT"]
}
/**
* Holds if `name` corresponds to a weak password hashing algorithm.
*/
predicate isWeakPasswordHashingAlgorithm(string name) { name = "EVPKDF" }
/**
* Holds if `name` corresponds to a stream cipher.
*/
predicate isStreamCipher(string name) { name = ["CHACHA", "RC4", "ARC4", "ARCFOUR", "RABBIT"] }

188
javascript/ql/lib/semmle/javascript/security/internal/SensitiveDataHeuristics.qll
View File

@@ -1,188 +0,0 @@
/**
* INTERNAL: Do not use.
*
* Provides classes and predicates for identifying strings that may indicate the presence of sensitive data.
* Such that we can share this logic across our CodeQL analysis of different languages.
*
* 'Sensitive' data in general is anything that should not be sent around in unencrypted form.
*/
/**
* A classification of different kinds of sensitive data:
*
* - secret: generic secret or trusted data;
* - id: a user name or other account information;
* - password: a password or authorization key;
* - certificate: a certificate.
* - private: private data such as credit card numbers
*
* While classifications are represented as strings, this should not be relied upon.
* Instead, use the predicates in `SensitiveDataClassification::` to work with
* classifications.
*/
class SensitiveDataClassification extends string {
SensitiveDataClassification() { this in ["secret", "id", "password", "certificate", "private"] }
}
/**
* Provides predicates to select the different kinds of sensitive data we support.
*/
module SensitiveDataClassification {
/** Gets the classification for secret or trusted data. */
SensitiveDataClassification secret() { result = "secret" }
/** Gets the classification for user names or other account information. */
SensitiveDataClassification id() { result = "id" }
/** Gets the classification for passwords or authorization keys. */
SensitiveDataClassification password() { result = "password" }
/** Gets the classification for certificates. */
SensitiveDataClassification certificate() { result = "certificate" }
/** Gets the classification for private data. */
SensitiveDataClassification private() { result = "private" }
}
/**
* INTERNAL: Do not use.
*
* Provides heuristics for identifying names related to sensitive information.
*/
module HeuristicNames {
/**
* Gets a regular expression that identifies strings that may indicate the presence of secret
* or trusted data.
*/
string maybeSecret() { result = "(?is).*((?<!is|is_)secret|(?<!un|un_|is|is_)trusted).*" }
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* user names or other account information.
*/
string maybeAccountInfo() {
result = "(?is).*acc(ou)?nt.*" or
result = "(?is).*(puid|user.?name|user.?id|session.?(id|key)).*" or
result = "(?s).*([uU]|^|_|[a-z](?=U))([uU][iI][dD]).*"
}
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* a password or an authorization key.
*/
string maybePassword() {
result = "(?is).*pass(wd|word|code|.?phrase)(?!.*question).*" or
result = "(?is).*(auth(entication|ori[sz]ation)?).?key.*"
}
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* a certificate.
*/
string maybeCertificate() { result = "(?is).*(cert)(?!.*(format|name|ification)).*" }
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* private data.
*/
string maybePrivate() {
result =
"(?is).*(" +
// Inspired by the list on https://cwe.mitre.org/data/definitions/359.html
// Government identifiers, such as Social Security Numbers
"social.?security|employer.?identification|national.?insurance|resident.?id|" +
"passport.?(num|no)|([_-]|\\b)ssn([_-]|\\b)|" +
// Contact information, such as home addresses
"post.?code|zip.?code|home.?addr|" +
// and telephone numbers
"(mob(ile)?|home).?(num|no|tel|phone)|(tel|fax|phone).?(num|no)|telephone|" +
"emergency.?contact|" +
// Geographic location - where the user is (or was)
"latitude|longitude|nationality|" +
// Financial data - such as credit card numbers, salary, bank accounts, and debts
"(credit|debit|bank|visa).?(card|num|no|acc(ou)?nt)|acc(ou)?nt.?(no|num|credit)|" +
"salary|billing|credit.?(rating|score)|([_-]|\\b)ccn([_-]|\\b)|" +
// Communications - e-mail addresses, private e-mail messages, SMS text messages, chat logs, etc.
// "e(mail|_mail)|" + // this seems too noisy
// Health - medical conditions, insurance status, prescription records
"birth.?da(te|y)|da(te|y).?(of.?)?birth|" +
"medical|(health|care).?plan|healthkit|appointment|prescription|" +
"blood.?(type|alcohol|glucose|pressure)|heart.?(rate|rhythm)|body.?(mass|fat)|" +
"menstrua|pregnan|insulin|inhaler|" +
// Relationships - work and family
"employ(er|ee)|spouse|maiden.?name" +
// ---
").*"
}
/**
* Gets a regular expression that identifies strings that may indicate the presence
* of sensitive data, with `classification` describing the kind of sensitive data involved.
*/
string maybeSensitiveRegexp(SensitiveDataClassification classification) {
result = maybeSecret() and classification = SensitiveDataClassification::secret()
or
result = maybeAccountInfo() and classification = SensitiveDataClassification::id()
or
result = maybePassword() and classification = SensitiveDataClassification::password()
or
result = maybeCertificate() and
classification = SensitiveDataClassification::certificate()
or
result = maybePrivate() and
classification = SensitiveDataClassification::private()
}
/**
* Gets a regular expression that identifies strings that may indicate the presence of data
* that is hashed or encrypted, and hence rendered non-sensitive, or contains special characters
* suggesting nouns within the string do not represent the meaning of the whole string (e.g. a URL or a SQL query).
*
* We also filter out common words like `certain` and `concert`, since otherwise these could
* be matched by the certificate regular expressions. Same for `accountable` (account), or
* `secretarial` (secret).
*/
string notSensitiveRegexp() {
result =
"(?is).*([^\\w$.-]|redact|censor|obfuscate|hash|md5|sha|random|((?<!un)(en))?(crypt|(?<!pass)code)|certain|concert|secretar|accountant|accountab).*"
}
/**
* Holds if `name` may indicate the presence of sensitive data, and `name` does not indicate that
* the data is in fact non-sensitive (for example since it is hashed or encrypted).
*
* That is, one of the regexps from `maybeSensitiveRegexp` matches `name` (with the given
* classification), and none of the regexps from `notSensitiveRegexp` matches `name`.
*/
bindingset[name]
predicate nameIndicatesSensitiveData(string name) {
exists(string combinedRegexp |
// Combine all the maybe-sensitive regexps into one using non-capturing groups and |.
combinedRegexp =
"(?:" + strictconcat(string r | r = maybeSensitiveRegexp(_) | r, ")|(?:") + ")"
|
name.regexpMatch(combinedRegexp)
) and
not name.regexpMatch(notSensitiveRegexp())
}
/**
* Holds if `name` may indicate the presence of sensitive data, and
* `name` does not indicate that the data is in fact non-sensitive (for example since
* it is hashed or encrypted). `classification` describes the kind of sensitive data
* involved.
*
* That is, one of the regexps from `maybeSensitiveRegexp` matches `name` (with the
* given classification), and none of the regexps from `notSensitiveRegexp` matches
* `name`.
*
* When the set of names is large, it's worth using `nameIndicatesSensitiveData/1` as a first
* pass, since that combines all the regexps into one, and should be faster. Then call this
* predicate to get the classification(s).
*/
bindingset[name]
predicate nameIndicatesSensitiveData(string name, SensitiveDataClassification classification) {
name.regexpMatch(maybeSensitiveRegexp(classification)) and
not name.regexpMatch(notSensitiveRegexp())
}
}

1
python/ql/lib/qlpack.yml
View File

@@ -6,6 +6,7 @@ extractor: python
library: true library: true
upgrades: upgrades upgrades: upgrades
dependencies: dependencies:
codeql/concepts: ${workspace}
codeql/dataflow: ${workspace} codeql/dataflow: ${workspace}
codeql/mad: ${workspace} codeql/mad: ${workspace}
codeql/regex: ${workspace} codeql/regex: ${workspace}

2
python/ql/lib/semmle/crypto/Crypto.qll
View File

@@ -1,3 +1,3 @@
/** DEPRECATED: Use `semmle.python.concepts.CryptoAlgorithms` instead. */ /** DEPRECATED: Use `semmle.python.concepts.CryptoAlgorithms` instead. */
import semmle.python.concepts.CryptoAlgorithms import codeql.concepts.CryptoAlgorithms

9
python/ql/lib/semmle/python/Concepts.qll
View File

@@ -6,11 +6,16 @@
private import python private import python
private import semmle.python.dataflow.new.DataFlow private import semmle.python.dataflow.new.DataFlow
private import semmle.python.dataflow.new.internal.DataFlowImplSpecific
private import semmle.python.dataflow.new.RemoteFlowSources private import semmle.python.dataflow.new.RemoteFlowSources
private import semmle.python.dataflow.new.TaintTracking private import semmle.python.dataflow.new.TaintTracking
private import semmle.python.Files
private import semmle.python.Frameworks private import semmle.python.Frameworks
private import semmle.python.security.internal.EncryptionKeySizes private import semmle.python.security.internal.EncryptionKeySizes
private import codeql.threatmodels.ThreatModels private import codeql.threatmodels.ThreatModels
private import codeql.concepts.ConceptsShared
private module ConceptsShared = ConceptsMake<Location, PythonDataFlow>;
/** /**
* A data flow source, for a specific threat-model. * A data flow source, for a specific threat-model.
@@ -1617,7 +1622,7 @@ module Http {
} }
} }
import semmle.python.internal.ConceptsShared::Http::Client as Client import ConceptsShared::Http::Client as Client
// TODO: investigate whether we should treat responses to client requests as // TODO: investigate whether we should treat responses to client requests as
// remote-flow-sources in general. // remote-flow-sources in general.
} }
@@ -1725,5 +1730,5 @@ module Cryptography {
} }
} }
import semmle.python.internal.ConceptsShared::Cryptography import ConceptsShared::Cryptography
} }

114
python/ql/lib/semmle/python/concepts/CryptoAlgorithms.qll
View File

@@ -1,117 +1,5 @@
/** /**
* Provides classes modeling cryptographic algorithms, separated into strong and weak variants. * Provides classes modeling cryptographic algorithms, separated into strong and weak variants.
*
* The classification into strong and weak are based on Wikipedia, OWASP and Google (2021).
*/ */
private import internal.CryptoAlgorithmNames import codeql.concepts.CryptoAlgorithms
/**
* A cryptographic algorithm.
*/
private newtype TCryptographicAlgorithm =
MkHashingAlgorithm(string name, boolean isWeak) {
isStrongHashingAlgorithm(name) and isWeak = false
or
isWeakHashingAlgorithm(name) and isWeak = true
} or
MkEncryptionAlgorithm(string name, boolean isWeak) {
isStrongEncryptionAlgorithm(name) and isWeak = false
or
isWeakEncryptionAlgorithm(name) and isWeak = true
} or
MkPasswordHashingAlgorithm(string name, boolean isWeak) {
isStrongPasswordHashingAlgorithm(name) and isWeak = false
or
isWeakPasswordHashingAlgorithm(name) and isWeak = true
}
/**
* Gets the most specific `CryptographicAlgorithm` that matches the given `name`.
* A matching algorithm is one where the name of the algorithm matches the start of name, with allowances made for different name formats.
* In the case that multiple `CryptographicAlgorithm`s match the given `name`, the algorithm(s) with the longest name will be selected. This is intended to select more specific versions of algorithms when multiple versions could match - for example "SHA3_224" matches against both "SHA3" and "SHA3224", but the latter is a more precise match.
*/
bindingset[name]
private CryptographicAlgorithm getBestAlgorithmForName(string name) {
result =
max(CryptographicAlgorithm algorithm |
algorithm.getName() =
[
name.toUpperCase(), // the full name
name.toUpperCase().regexpCapture("^([\\w]+)(?:-.*)?$", 1), // the name prior to any dashes or spaces
name.toUpperCase().regexpCapture("^([A-Z0-9]+)(?:(-|_).*)?$", 1) // the name prior to any dashes, spaces, or underscores
].regexpReplaceAll("[-_ ]", "") // strip dashes, underscores, and spaces
|
algorithm order by algorithm.getName().length()
)
}
/**
* A cryptographic algorithm.
*/
abstract class CryptographicAlgorithm extends TCryptographicAlgorithm {
/** Gets a textual representation of this element. */
string toString() { result = this.getName() }
/**
* Gets the normalized name of this algorithm (upper-case, no spaces, dashes or underscores).
*/
abstract string getName();
/**
* Holds if the name of this algorithm is the most specific match for `name`.
* This predicate matches quite liberally to account for different ways of formatting algorithm names, e.g. using dashes, underscores, or spaces as separators, including or not including block modes of operation, etc.
*/
bindingset[name]
predicate matchesName(string name) { this = getBestAlgorithmForName(name) }
/**
* Holds if this algorithm is weak.
*/
abstract predicate isWeak();
}
/**
* A hashing algorithm such as `MD5` or `SHA512`.
*/
class HashingAlgorithm extends MkHashingAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
HashingAlgorithm() { this = MkHashingAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
}
/**
* An encryption algorithm such as `DES` or `AES512`.
*/
class EncryptionAlgorithm extends MkEncryptionAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
EncryptionAlgorithm() { this = MkEncryptionAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
/** Holds if this algorithm is a stream cipher. */
predicate isStreamCipher() { isStreamCipher(name) }
}
/**
* A password hashing algorithm such as `PBKDF2` or `SCRYPT`.
*/
class PasswordHashingAlgorithm extends MkPasswordHashingAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
PasswordHashingAlgorithm() { this = MkPasswordHashingAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
}

2
python/ql/lib/semmle/python/dataflow/new/SensitiveDataSources.qll
View File

@@ -7,7 +7,7 @@ private import python
private import semmle.python.dataflow.new.DataFlow private import semmle.python.dataflow.new.DataFlow
// Need to import `semmle.python.Frameworks` since frameworks can extend `SensitiveDataSource::Range` // Need to import `semmle.python.Frameworks` since frameworks can extend `SensitiveDataSource::Range`
private import semmle.python.Frameworks private import semmle.python.Frameworks
private import semmle.python.security.internal.SensitiveDataHeuristics as SensitiveDataHeuristics private import codeql.concepts.internal.SensitiveDataHeuristics as SensitiveDataHeuristics
private import semmle.python.ApiGraphs private import semmle.python.ApiGraphs
// We export these explicitly, so we don't also export the `HeuristicNames` module. // We export these explicitly, so we don't also export the `HeuristicNames` module.

10
python/ql/lib/semmle/python/frameworks/Aiohttp.qll
View File

@@ -758,7 +758,7 @@ module AiohttpClientModel {
private API::Node instance() { result = classRef().getReturn() } private API::Node instance() { result = classRef().getReturn() }
/** A method call on a ClientSession that sends off a request */ /** A method call on a ClientSession that sends off a request */
private class OutgoingRequestCall extends Http::Client::Request::Range, API::CallNode { private class OutgoingRequestCall extends Http::Client::Request::Range instanceof API::CallNode {
string methodName; string methodName;
OutgoingRequestCall() { OutgoingRequestCall() {
@@ -767,13 +767,13 @@ module AiohttpClientModel {
} }
override DataFlow::Node getAUrlPart() { override DataFlow::Node getAUrlPart() {
result = this.getArgByName("url") result = super.getArgByName("url")
or or
methodName in [Http::httpVerbLower(), "ws_connect"] and methodName in [Http::httpVerbLower(), "ws_connect"] and
result = this.getArg(0) result = super.getArg(0)
or or
methodName = "request" and methodName = "request" and
result = this.getArg(1) result = super.getArg(1)
} }
override string getFramework() { result = "aiohttp.ClientSession" } override string getFramework() { result = "aiohttp.ClientSession" }
@@ -781,7 +781,7 @@ module AiohttpClientModel {
override predicate disablesCertificateValidation( override predicate disablesCertificateValidation(
DataFlow::Node disablingNode, DataFlow::Node argumentOrigin DataFlow::Node disablingNode, DataFlow::Node argumentOrigin
) { ) {
exists(API::Node param | param = this.getKeywordParameter(["ssl", "verify_ssl"]) | exists(API::Node param | param = super.getKeywordParameter(["ssl", "verify_ssl"]) |
disablingNode = param.asSink() and disablingNode = param.asSink() and
argumentOrigin = param.getAValueReachingSink() and argumentOrigin = param.getAValueReachingSink() and
// aiohttp.client treats `None` as the default and all other "falsey" values as `False`. // aiohttp.client treats `None` as the default and all other "falsey" values as `False`.

33
python/ql/lib/semmle/python/frameworks/Cryptodome.qll
View File

@@ -107,8 +107,7 @@ private module CryptodomeModel {
/** /**
* A cryptographic operation on an instance from the `Cipher` subpackage of `Cryptodome`/`Crypto`. * A cryptographic operation on an instance from the `Cipher` subpackage of `Cryptodome`/`Crypto`.
*/ */
class CryptodomeGenericCipherOperation extends Cryptography::CryptographicOperation::Range, class CryptodomeGenericCipherOperation extends Cryptography::CryptographicOperation::Range instanceof DataFlow::CallCfgNode
DataFlow::CallCfgNode
{ {
string methodName; string methodName;
string cipherName; string cipherName;
@@ -134,31 +133,31 @@ private module CryptodomeModel {
override DataFlow::Node getAnInput() { override DataFlow::Node getAnInput() {
methodName = "encrypt" and methodName = "encrypt" and
result in [this.getArg(0), this.getArgByName(["message", "plaintext"])] result in [super.getArg(0), super.getArgByName(["message", "plaintext"])]
or or
methodName = "decrypt" and methodName = "decrypt" and
result in [this.getArg(0), this.getArgByName("ciphertext")] result in [super.getArg(0), super.getArgByName("ciphertext")]
or or
// for the following methods, method signatures can be found in // for the following methods, method signatures can be found in
// https://pycryptodome.readthedocs.io/en/latest/src/cipher/modern.html // https://pycryptodome.readthedocs.io/en/latest/src/cipher/modern.html
methodName = "update" and methodName = "update" and
result in [this.getArg(0), this.getArgByName("data")] result in [super.getArg(0), super.getArgByName("data")]
or or
// although `mac_tag` is used as the parameter name in the spec above, some implementations use `received_mac_tag`, for an example, see // although `mac_tag` is used as the parameter name in the spec above, some implementations use `received_mac_tag`, for an example, see
// https://github.com/Legrandin/pycryptodome/blob/5dace638b70ac35bb5d9b565f3e75f7869c9d851/lib/Crypto/Cipher/ChaCha20_Poly1305.py#L207 // https://github.com/Legrandin/pycryptodome/blob/5dace638b70ac35bb5d9b565f3e75f7869c9d851/lib/Crypto/Cipher/ChaCha20_Poly1305.py#L207
methodName = "verify" and methodName = "verify" and
result in [this.getArg(0), this.getArgByName(["mac_tag", "received_mac_tag"])] result in [super.getArg(0), super.getArgByName(["mac_tag", "received_mac_tag"])]
or or
methodName = "hexverify" and methodName = "hexverify" and
result in [this.getArg(0), this.getArgByName("mac_tag_hex")] result in [super.getArg(0), super.getArgByName("mac_tag_hex")]
or or
methodName = "encrypt_and_digest" and methodName = "encrypt_and_digest" and
result in [this.getArg(0), this.getArgByName("plaintext")] result in [super.getArg(0), super.getArgByName("plaintext")]
or or
methodName = "decrypt_and_verify" and methodName = "decrypt_and_verify" and
result in [ result in [
this.getArg(0), this.getArgByName("ciphertext"), this.getArg(1), super.getArg(0), super.getArgByName("ciphertext"), super.getArg(1),
this.getArgByName("mac_tag") super.getArgByName("mac_tag")
] ]
} }
@@ -180,8 +179,7 @@ private module CryptodomeModel {
/** /**
* A cryptographic operation on an instance from the `Signature` subpackage of `Cryptodome`/`Crypto`. * A cryptographic operation on an instance from the `Signature` subpackage of `Cryptodome`/`Crypto`.
*/ */
class CryptodomeGenericSignatureOperation extends Cryptography::CryptographicOperation::Range, class CryptodomeGenericSignatureOperation extends Cryptography::CryptographicOperation::Range instanceof DataFlow::CallCfgNode
DataFlow::CallCfgNode
{ {
API::CallNode newCall; API::CallNode newCall;
string methodName; string methodName;
@@ -206,13 +204,13 @@ private module CryptodomeModel {
override DataFlow::Node getAnInput() { override DataFlow::Node getAnInput() {
methodName = "sign" and methodName = "sign" and
result in [this.getArg(0), this.getArgByName("msg_hash")] // Cryptodome.Hash instance result in [super.getArg(0), super.getArgByName("msg_hash")] // Cryptodome.Hash instance
or or
methodName = "verify" and methodName = "verify" and
( (
result in [this.getArg(0), this.getArgByName("msg_hash")] // Cryptodome.Hash instance result in [super.getArg(0), super.getArgByName("msg_hash")] // Cryptodome.Hash instance
or or
result in [this.getArg(1), this.getArgByName("signature")] result in [super.getArg(1), super.getArgByName("signature")]
) )
} }
@@ -222,8 +220,7 @@ private module CryptodomeModel {
/** /**
* A cryptographic operation on an instance from the `Hash` subpackage of `Cryptodome`/`Crypto`. * A cryptographic operation on an instance from the `Hash` subpackage of `Cryptodome`/`Crypto`.
*/ */
class CryptodomeGenericHashOperation extends Cryptography::CryptographicOperation::Range, class CryptodomeGenericHashOperation extends Cryptography::CryptographicOperation::Range instanceof DataFlow::CallCfgNode
DataFlow::CallCfgNode
{ {
API::CallNode newCall; API::CallNode newCall;
string hashName; string hashName;
@@ -244,7 +241,7 @@ private module CryptodomeModel {
override Cryptography::CryptographicAlgorithm getAlgorithm() { result.matchesName(hashName) } override Cryptography::CryptographicAlgorithm getAlgorithm() { result.matchesName(hashName) }
override DataFlow::Node getAnInput() { result in [this.getArg(0), this.getArgByName("data")] } override DataFlow::Node getAnInput() { result in [super.getArg(0), super.getArgByName("data")] }
override Cryptography::BlockMode getBlockMode() { none() } override Cryptography::BlockMode getBlockMode() { none() }
} }

14
python/ql/lib/semmle/python/frameworks/Cryptography.qll
View File

@@ -206,8 +206,7 @@ private module CryptographyModel {
/** /**
* An encrypt or decrypt operation from `cryptography.hazmat.primitives.ciphers`. * An encrypt or decrypt operation from `cryptography.hazmat.primitives.ciphers`.
*/ */
class CryptographyGenericCipherOperation extends Cryptography::CryptographicOperation::Range, class CryptographyGenericCipherOperation extends Cryptography::CryptographicOperation::Range instanceof DataFlow::MethodCallNode
DataFlow::MethodCallNode
{ {
API::CallNode init; API::CallNode init;
string algorithmName; string algorithmName;
@@ -225,7 +224,9 @@ private module CryptographyModel {
result.matchesName(algorithmName) result.matchesName(algorithmName)
} }
override DataFlow::Node getAnInput() { result in [this.getArg(0), this.getArgByName("data")] } override DataFlow::Node getAnInput() {
result in [super.getArg(0), super.getArgByName("data")]
}
override Cryptography::BlockMode getBlockMode() { result = modeName } override Cryptography::BlockMode getBlockMode() { result = modeName }
} }
@@ -263,8 +264,7 @@ private module CryptographyModel {
/** /**
* An hashing operation from `cryptography.hazmat.primitives.hashes`. * An hashing operation from `cryptography.hazmat.primitives.hashes`.
*/ */
class CryptographyGenericHashOperation extends Cryptography::CryptographicOperation::Range, class CryptographyGenericHashOperation extends Cryptography::CryptographicOperation::Range instanceof DataFlow::MethodCallNode
DataFlow::MethodCallNode
{ {
API::CallNode init; API::CallNode init;
string algorithmName; string algorithmName;
@@ -280,7 +280,9 @@ private module CryptographyModel {
result.matchesName(algorithmName) result.matchesName(algorithmName)
} }
override DataFlow::Node getAnInput() { result in [this.getArg(0), this.getArgByName("data")] } override DataFlow::Node getAnInput() {
result in [super.getArg(0), super.getArgByName("data")]
}
override Cryptography::BlockMode getBlockMode() { none() } override Cryptography::BlockMode getBlockMode() { none() }
} }

21
python/ql/lib/semmle/python/frameworks/Httpx.qll
View File

@@ -26,7 +26,7 @@ module HttpxModel {
* *
* See https://www.python-httpx.org/api/ * See https://www.python-httpx.org/api/
*/ */
private class RequestCall extends Http::Client::Request::Range, API::CallNode { private class RequestCall extends Http::Client::Request::Range instanceof API::CallNode {
string methodName; string methodName;
RequestCall() { RequestCall() {
@@ -35,11 +35,11 @@ module HttpxModel {
} }
override DataFlow::Node getAUrlPart() { override DataFlow::Node getAUrlPart() {
result = this.getArgByName("url") result = super.getArgByName("url")
or or
if methodName in ["request", "stream"] if methodName in ["request", "stream"]
then result = this.getArg(1) then result = super.getArg(1)
else result = this.getArg(0) else result = super.getArg(0)
} }
override string getFramework() { result = "httpx" } override string getFramework() { result = "httpx" }
@@ -47,8 +47,8 @@ module HttpxModel {
override predicate disablesCertificateValidation( override predicate disablesCertificateValidation(
DataFlow::Node disablingNode, DataFlow::Node argumentOrigin DataFlow::Node disablingNode, DataFlow::Node argumentOrigin
) { ) {
disablingNode = this.getKeywordParameter("verify").asSink() and disablingNode = super.getKeywordParameter("verify").asSink() and
argumentOrigin = this.getKeywordParameter("verify").getAValueReachingSink() and argumentOrigin = super.getKeywordParameter("verify").getAValueReachingSink() and
// unlike `requests`, httpx treats `None` as turning off verify (and not as the default) // unlike `requests`, httpx treats `None` as turning off verify (and not as the default)
argumentOrigin.asExpr().(ImmutableLiteral).booleanValue() = false argumentOrigin.asExpr().(ImmutableLiteral).booleanValue() = false
// TODO: Handling of insecure SSLContext passed to verify argument // TODO: Handling of insecure SSLContext passed to verify argument
@@ -69,7 +69,8 @@ module HttpxModel {
} }
/** A method call on a Client that sends off a request */ /** A method call on a Client that sends off a request */
private class OutgoingRequestCall extends Http::Client::Request::Range, DataFlow::CallCfgNode { private class OutgoingRequestCall extends Http::Client::Request::Range instanceof DataFlow::CallCfgNode
{
string methodName; string methodName;
OutgoingRequestCall() { OutgoingRequestCall() {
@@ -78,11 +79,11 @@ module HttpxModel {
} }
override DataFlow::Node getAUrlPart() { override DataFlow::Node getAUrlPart() {
result = this.getArgByName("url") result = super.getArgByName("url")
or or
if methodName in ["request", "stream"] if methodName in ["request", "stream"]
then result = this.getArg(1) then result = super.getArg(1)
else result = this.getArg(0) else result = super.getArg(0)
} }
override string getFramework() { result = "httpx.[Async]Client" } override string getFramework() { result = "httpx.[Async]Client" }

4
python/ql/lib/semmle/python/frameworks/Libtaxii.qll
View File

@@ -22,13 +22,13 @@ private module Libtaxii {
* A call to `libtaxii.common.parse`. * A call to `libtaxii.common.parse`.
* When the `allow_url` parameter value is set to `True`, there is an SSRF vulnerability.. * When the `allow_url` parameter value is set to `True`, there is an SSRF vulnerability..
*/ */
private class ParseCall extends Http::Client::Request::Range, DataFlow::CallCfgNode { private class ParseCall extends Http::Client::Request::Range instanceof DataFlow::CallCfgNode {
ParseCall() { ParseCall() {
this = API::moduleImport("libtaxii").getMember("common").getMember("parse").getACall() and this = API::moduleImport("libtaxii").getMember("common").getMember("parse").getACall() and
this.getArgByName("allow_url").getALocalSource().asExpr() = any(True t) this.getArgByName("allow_url").getALocalSource().asExpr() = any(True t)
} }
override DataFlow::Node getAUrlPart() { result in [this.getArg(0), this.getArgByName("s")] } override DataFlow::Node getAUrlPart() { result in [super.getArg(0), super.getArgByName("s")] }
override string getFramework() { result = "libtaxii.common.parse" } override string getFramework() { result = "libtaxii.common.parse" }

15
python/ql/lib/semmle/python/frameworks/Pycurl.qll
View File

@@ -52,14 +52,15 @@ module Pycurl {
* *
* See http://pycurl.io/docs/latest/curlobject.html#pycurl.Curl.setopt. * See http://pycurl.io/docs/latest/curlobject.html#pycurl.Curl.setopt.
*/ */
private class OutgoingRequestCall extends Http::Client::Request::Range, DataFlow::CallCfgNode { private class OutgoingRequestCall extends Http::Client::Request::Range instanceof DataFlow::CallCfgNode
{
OutgoingRequestCall() { OutgoingRequestCall() {
this = setopt().getACall() and this = setopt().getACall() and
this.getArg(0).asCfgNode().(AttrNode).getName() = "URL" this.getArg(0).asCfgNode().(AttrNode).getName() = "URL"
} }
override DataFlow::Node getAUrlPart() { override DataFlow::Node getAUrlPart() {
result in [this.getArg(1), this.getArgByName("value")] result in [super.getArg(1), super.getArgByName("value")]
} }
override string getFramework() { result = "pycurl.Curl" } override string getFramework() { result = "pycurl.Curl" }
@@ -77,7 +78,7 @@ module Pycurl {
* *
* See http://pycurl.io/docs/latest/curlobject.html#pycurl.Curl.setopt. * See http://pycurl.io/docs/latest/curlobject.html#pycurl.Curl.setopt.
*/ */
private class CurlSslCall extends Http::Client::Request::Range, DataFlow::CallCfgNode { private class CurlSslCall extends Http::Client::Request::Range instanceof DataFlow::CallCfgNode {
CurlSslCall() { CurlSslCall() {
this = setopt().getACall() and this = setopt().getACall() and
this.getArg(0).asCfgNode().(AttrNode).getName() = ["SSL_VERIFYPEER", "SSL_VERIFYHOST"] this.getArg(0).asCfgNode().(AttrNode).getName() = ["SSL_VERIFYPEER", "SSL_VERIFYHOST"]
@@ -90,13 +91,13 @@ module Pycurl {
override predicate disablesCertificateValidation( override predicate disablesCertificateValidation(
DataFlow::Node disablingNode, DataFlow::Node argumentOrigin DataFlow::Node disablingNode, DataFlow::Node argumentOrigin
) { ) {
sslverifypeer().getAValueReachableFromSource() = this.getArg(0) and sslverifypeer().getAValueReachableFromSource() = super.getArg(0) and
( (
this.getArg(1).asExpr().(IntegerLiteral).getValue() = 0 super.getArg(1).asExpr().(IntegerLiteral).getValue() = 0
or or
this.getArg(1).asExpr().(BooleanLiteral).booleanValue() = false super.getArg(1).asExpr().(BooleanLiteral).booleanValue() = false
) and ) and
(disablingNode = this and argumentOrigin = this.getArg(1)) (disablingNode = this and argumentOrigin = super.getArg(1))
} }
} }
} }

12
python/ql/lib/semmle/python/frameworks/Requests.qll
View File

@@ -29,7 +29,7 @@ module Requests {
* *
* See https://requests.readthedocs.io/en/latest/api/#requests.request * See https://requests.readthedocs.io/en/latest/api/#requests.request
*/ */
private class OutgoingRequestCall extends Http::Client::Request::Range, API::CallNode { private class OutgoingRequestCall extends Http::Client::Request::Range instanceof API::CallNode {
string methodName; string methodName;
OutgoingRequestCall() { OutgoingRequestCall() {
@@ -50,20 +50,20 @@ module Requests {
} }
override DataFlow::Node getAUrlPart() { override DataFlow::Node getAUrlPart() {
result = this.getArgByName("url") result = super.getArgByName("url")
or or
not methodName = "request" and not methodName = "request" and
result = this.getArg(0) result = super.getArg(0)
or or
methodName = "request" and methodName = "request" and
result = this.getArg(1) result = super.getArg(1)
} }
override predicate disablesCertificateValidation( override predicate disablesCertificateValidation(
DataFlow::Node disablingNode, DataFlow::Node argumentOrigin DataFlow::Node disablingNode, DataFlow::Node argumentOrigin
) { ) {
disablingNode = this.getKeywordParameter("verify").asSink() and disablingNode = super.getKeywordParameter("verify").asSink() and
argumentOrigin = this.getKeywordParameter("verify").getAValueReachingSink() and argumentOrigin = super.getKeywordParameter("verify").getAValueReachingSink() and
// requests treats `None` as the default and all other "falsey" values as `False`. // requests treats `None` as the default and all other "falsey" values as `False`.
argumentOrigin.asExpr().(ImmutableLiteral).booleanValue() = false and argumentOrigin.asExpr().(ImmutableLiteral).booleanValue() = false and
not argumentOrigin.asExpr() instanceof None not argumentOrigin.asExpr() instanceof None

38
python/ql/lib/semmle/python/frameworks/Rsa.qll
View File

@@ -34,7 +34,8 @@ private module Rsa {
* *
* See https://stuvel.eu/python-rsa-doc/reference.html#rsa.encrypt * See https://stuvel.eu/python-rsa-doc/reference.html#rsa.encrypt
*/ */
class RsaEncryptCall extends Cryptography::CryptographicOperation::Range, DataFlow::CallCfgNode { class RsaEncryptCall extends Cryptography::CryptographicOperation::Range instanceof DataFlow::CallCfgNode
{
RsaEncryptCall() { this = API::moduleImport("rsa").getMember("encrypt").getACall() } RsaEncryptCall() { this = API::moduleImport("rsa").getMember("encrypt").getACall() }
override DataFlow::Node getInitialization() { result = this } override DataFlow::Node getInitialization() { result = this }
@@ -42,7 +43,7 @@ private module Rsa {
override Cryptography::CryptographicAlgorithm getAlgorithm() { result.getName() = "RSA" } override Cryptography::CryptographicAlgorithm getAlgorithm() { result.getName() = "RSA" }
override DataFlow::Node getAnInput() { override DataFlow::Node getAnInput() {
result in [this.getArg(0), this.getArgByName("message")] result in [super.getArg(0), super.getArgByName("message")]
} }
override Cryptography::BlockMode getBlockMode() { none() } override Cryptography::BlockMode getBlockMode() { none() }
@@ -53,14 +54,17 @@ private module Rsa {
* *
* See https://stuvel.eu/python-rsa-doc/reference.html#rsa.decrypt * See https://stuvel.eu/python-rsa-doc/reference.html#rsa.decrypt
*/ */
class RsaDecryptCall extends Cryptography::CryptographicOperation::Range, DataFlow::CallCfgNode { class RsaDecryptCall extends Cryptography::CryptographicOperation::Range instanceof DataFlow::CallCfgNode
{
RsaDecryptCall() { this = API::moduleImport("rsa").getMember("decrypt").getACall() } RsaDecryptCall() { this = API::moduleImport("rsa").getMember("decrypt").getACall() }
override DataFlow::Node getInitialization() { result = this } override DataFlow::Node getInitialization() { result = this }
override Cryptography::CryptographicAlgorithm getAlgorithm() { result.getName() = "RSA" } override Cryptography::CryptographicAlgorithm getAlgorithm() { result.getName() = "RSA" }
override DataFlow::Node getAnInput() { result in [this.getArg(0), this.getArgByName("crypto")] } override DataFlow::Node getAnInput() {
result in [super.getArg(0), super.getArgByName("crypto")]
}
override Cryptography::BlockMode getBlockMode() { none() } override Cryptography::BlockMode getBlockMode() { none() }
} }
@@ -70,7 +74,8 @@ private module Rsa {
* *
* See https://stuvel.eu/python-rsa-doc/reference.html#rsa.sign * See https://stuvel.eu/python-rsa-doc/reference.html#rsa.sign
*/ */
class RsaSignCall extends Cryptography::CryptographicOperation::Range, DataFlow::CallCfgNode { class RsaSignCall extends Cryptography::CryptographicOperation::Range instanceof DataFlow::CallCfgNode
{
RsaSignCall() { this = API::moduleImport("rsa").getMember("sign").getACall() } RsaSignCall() { this = API::moduleImport("rsa").getMember("sign").getACall() }
override DataFlow::Node getInitialization() { result = this } override DataFlow::Node getInitialization() { result = this }
@@ -81,14 +86,14 @@ private module Rsa {
or or
// hashing part // hashing part
exists(StringLiteral str, DataFlow::Node hashNameArg | exists(StringLiteral str, DataFlow::Node hashNameArg |
hashNameArg in [this.getArg(2), this.getArgByName("hash_method")] and hashNameArg in [super.getArg(2), super.getArgByName("hash_method")] and
DataFlow::exprNode(str) = hashNameArg.getALocalSource() and DataFlow::exprNode(str) = hashNameArg.getALocalSource() and
result.matchesName(str.getText()) result.matchesName(str.getText())
) )
} }
override DataFlow::Node getAnInput() { override DataFlow::Node getAnInput() {
result in [this.getArg(0), this.getArgByName("message")] result in [super.getArg(0), super.getArgByName("message")]
} }
override Cryptography::BlockMode getBlockMode() { none() } override Cryptography::BlockMode getBlockMode() { none() }
@@ -99,7 +104,8 @@ private module Rsa {
* *
* See https://stuvel.eu/python-rsa-doc/reference.html#rsa.verify * See https://stuvel.eu/python-rsa-doc/reference.html#rsa.verify
*/ */
class RsaVerifyCall extends Cryptography::CryptographicOperation::Range, DataFlow::CallCfgNode { class RsaVerifyCall extends Cryptography::CryptographicOperation::Range instanceof DataFlow::CallCfgNode
{
RsaVerifyCall() { this = API::moduleImport("rsa").getMember("verify").getACall() } RsaVerifyCall() { this = API::moduleImport("rsa").getMember("verify").getACall() }
override DataFlow::Node getInitialization() { result = this } override DataFlow::Node getInitialization() { result = this }
@@ -111,9 +117,9 @@ private module Rsa {
} }
override DataFlow::Node getAnInput() { override DataFlow::Node getAnInput() {
result in [this.getArg(0), this.getArgByName("message")] result in [super.getArg(0), super.getArgByName("message")]
or or
result in [this.getArg(1), this.getArgByName("signature")] result in [super.getArg(1), super.getArgByName("signature")]
} }
override Cryptography::BlockMode getBlockMode() { none() } override Cryptography::BlockMode getBlockMode() { none() }
@@ -124,8 +130,7 @@ private module Rsa {
* *
* See https://stuvel.eu/python-rsa-doc/reference.html#rsa.compute_hash * See https://stuvel.eu/python-rsa-doc/reference.html#rsa.compute_hash
*/ */
class RsaComputeHashCall extends Cryptography::CryptographicOperation::Range, class RsaComputeHashCall extends Cryptography::CryptographicOperation::Range instanceof DataFlow::CallCfgNode
DataFlow::CallCfgNode
{ {
RsaComputeHashCall() { this = API::moduleImport("rsa").getMember("compute_hash").getACall() } RsaComputeHashCall() { this = API::moduleImport("rsa").getMember("compute_hash").getACall() }
@@ -133,14 +138,14 @@ private module Rsa {
override Cryptography::CryptographicAlgorithm getAlgorithm() { override Cryptography::CryptographicAlgorithm getAlgorithm() {
exists(StringLiteral str, DataFlow::Node hashNameArg | exists(StringLiteral str, DataFlow::Node hashNameArg |
hashNameArg in [this.getArg(1), this.getArgByName("method_name")] and hashNameArg in [super.getArg(1), super.getArgByName("method_name")] and
DataFlow::exprNode(str) = hashNameArg.getALocalSource() and DataFlow::exprNode(str) = hashNameArg.getALocalSource() and
result.matchesName(str.getText()) result.matchesName(str.getText())
) )
} }
override DataFlow::Node getAnInput() { override DataFlow::Node getAnInput() {
result in [this.getArg(0), this.getArgByName("message")] result in [super.getArg(0), super.getArgByName("message")]
} }
override Cryptography::BlockMode getBlockMode() { none() } override Cryptography::BlockMode getBlockMode() { none() }
@@ -151,7 +156,8 @@ private module Rsa {
* *
* See https://stuvel.eu/python-rsa-doc/reference.html#rsa.sign_hash * See https://stuvel.eu/python-rsa-doc/reference.html#rsa.sign_hash
*/ */
class RsaSignHashCall extends Cryptography::CryptographicOperation::Range, DataFlow::CallCfgNode { class RsaSignHashCall extends Cryptography::CryptographicOperation::Range instanceof DataFlow::CallCfgNode
{
RsaSignHashCall() { this = API::moduleImport("rsa").getMember("sign_hash").getACall() } RsaSignHashCall() { this = API::moduleImport("rsa").getMember("sign_hash").getACall() }
override DataFlow::Node getInitialization() { result = this } override DataFlow::Node getInitialization() { result = this }
@@ -159,7 +165,7 @@ private module Rsa {
override Cryptography::CryptographicAlgorithm getAlgorithm() { result.getName() = "RSA" } override Cryptography::CryptographicAlgorithm getAlgorithm() { result.getName() = "RSA" }
override DataFlow::Node getAnInput() { override DataFlow::Node getAnInput() {
result in [this.getArg(0), this.getArgByName("hash_value")] result in [super.getArg(0), super.getArgByName("hash_value")]
} }
override Cryptography::BlockMode getBlockMode() { none() } override Cryptography::BlockMode getBlockMode() { none() }

54
python/ql/lib/semmle/python/frameworks/Stdlib.qll
View File

@@ -2385,15 +2385,16 @@ module StdlibPrivate {
} }
/** A method call on a HttpConnection that sends off a request */ /** A method call on a HttpConnection that sends off a request */
private class RequestCall extends Http::Client::Request::Range, DataFlow::MethodCallNode { private class RequestCall extends Http::Client::Request::Range instanceof DataFlow::MethodCallNode
{
RequestCall() { this.calls(instance(_), ["request", "_send_request", "putrequest"]) } RequestCall() { this.calls(instance(_), ["request", "_send_request", "putrequest"]) }
DataFlow::Node getUrlArg() { result in [this.getArg(1), this.getArgByName("url")] } DataFlow::Node getUrlArg() { result in [super.getArg(1), super.getArgByName("url")] }
override DataFlow::Node getAUrlPart() { override DataFlow::Node getAUrlPart() {
result = this.getUrlArg() result = this.getUrlArg()
or or
this.getObject() = instance(result) super.getObject() = instance(result)
} }
override string getFramework() { result = "http.client.HTTP[S]Connection" } override string getFramework() { result = "http.client.HTTP[S]Connection" }
@@ -2430,7 +2431,8 @@ module StdlibPrivate {
// a request method // a request method
exists(RequestCall call | exists(RequestCall call |
nodeFrom = call.getUrlArg() and nodeFrom = call.getUrlArg() and
nodeTo.(DataFlow::PostUpdateNode).getPreUpdateNode() = call.getObject() nodeTo.(DataFlow::PostUpdateNode).getPreUpdateNode() =
call.(DataFlow::MethodCallNode).getObject()
) )
or or
// `getresponse` call // `getresponse` call
@@ -2797,7 +2799,7 @@ module StdlibPrivate {
/** /**
* A hashing operation by supplying initial data when calling the `hashlib.new` function. * A hashing operation by supplying initial data when calling the `hashlib.new` function.
*/ */
class HashlibNewCall extends Cryptography::CryptographicOperation::Range, API::CallNode { class HashlibNewCall extends Cryptography::CryptographicOperation::Range instanceof API::CallNode {
string hashName; string hashName;
HashlibNewCall() { HashlibNewCall() {
@@ -2810,7 +2812,7 @@ module StdlibPrivate {
override Cryptography::CryptographicAlgorithm getAlgorithm() { result.matchesName(hashName) } override Cryptography::CryptographicAlgorithm getAlgorithm() { result.matchesName(hashName) }
override DataFlow::Node getAnInput() { result = this.getParameter(1, "data").asSink() } override DataFlow::Node getAnInput() { result = super.getParameter(1, "data").asSink() }
override Cryptography::BlockMode getBlockMode() { none() } override Cryptography::BlockMode getBlockMode() { none() }
} }
@@ -2818,7 +2820,8 @@ module StdlibPrivate {
/** /**
* A hashing operation by using the `update` method on the result of calling the `hashlib.new` function. * A hashing operation by using the `update` method on the result of calling the `hashlib.new` function.
*/ */
class HashlibNewUpdateCall extends Cryptography::CryptographicOperation::Range, API::CallNode { class HashlibNewUpdateCall extends Cryptography::CryptographicOperation::Range instanceof API::CallNode
{
API::CallNode init; API::CallNode init;
string hashName; string hashName;
@@ -2831,7 +2834,7 @@ module StdlibPrivate {
override Cryptography::CryptographicAlgorithm getAlgorithm() { result.matchesName(hashName) } override Cryptography::CryptographicAlgorithm getAlgorithm() { result.matchesName(hashName) }
override DataFlow::Node getAnInput() { result = this.getArg(0) } override DataFlow::Node getAnInput() { result = super.getArg(0) }
override Cryptography::BlockMode getBlockMode() { none() } override Cryptography::BlockMode getBlockMode() { none() }
} }
@@ -2848,8 +2851,7 @@ module StdlibPrivate {
* (such as `hashlib.md5`). `hashlib.new` is not included, since it is handled by * (such as `hashlib.md5`). `hashlib.new` is not included, since it is handled by
* `HashlibNewCall` and `HashlibNewUpdateCall`. * `HashlibNewCall` and `HashlibNewUpdateCall`.
*/ */
abstract class HashlibGenericHashOperation extends Cryptography::CryptographicOperation::Range, abstract class HashlibGenericHashOperation extends Cryptography::CryptographicOperation::Range instanceof DataFlow::CallCfgNode
DataFlow::CallCfgNode
{ {
string hashName; string hashName;
API::Node hashClass; API::Node hashClass;
@@ -2876,7 +2878,7 @@ module StdlibPrivate {
override DataFlow::Node getInitialization() { result = init } override DataFlow::Node getInitialization() { result = init }
override DataFlow::Node getAnInput() { result = this.getArg(0) } override DataFlow::Node getAnInput() { result = this.(DataFlow::CallCfgNode).getArg(0) }
} }
/** /**
@@ -2888,24 +2890,28 @@ module StdlibPrivate {
// we only want to model calls to classes such as `hashlib.md5()` if initial data // we only want to model calls to classes such as `hashlib.md5()` if initial data
// is passed as an argument // is passed as an argument
this = hashClass.getACall() and this = hashClass.getACall() and
exists([this.getArg(0), this.getArgByName("string")]) exists(
[
this.(DataFlow::CallCfgNode).getArg(0),
this.(DataFlow::CallCfgNode).getArgByName("string")
]
)
} }
override DataFlow::Node getInitialization() { result = this } override DataFlow::Node getInitialization() { result = this }
override DataFlow::Node getAnInput() { override DataFlow::Node getAnInput() {
result = this.getArg(0) result = this.(DataFlow::CallCfgNode).getArg(0)
or or
// in Python 3.9, you are allowed to use `hashlib.md5(string=<bytes-like>)`. // in Python 3.9, you are allowed to use `hashlib.md5(string=<bytes-like>)`.
result = this.getArgByName("string") result = this.(DataFlow::CallCfgNode).getArgByName("string")
} }
} }
// --------------------------------------------------------------------------- // ---------------------------------------------------------------------------
// hmac // hmac
// --------------------------------------------------------------------------- // ---------------------------------------------------------------------------
abstract class HmacCryptographicOperation extends Cryptography::CryptographicOperation::Range, abstract class HmacCryptographicOperation extends Cryptography::CryptographicOperation::Range instanceof API::CallNode
API::CallNode
{ {
abstract API::Node getDigestArg(); abstract API::Node getDigestArg();
@@ -2937,14 +2943,16 @@ module StdlibPrivate {
HmacNewCall() { HmacNewCall() {
this = getHmacConstructorCall(digestArg) and this = getHmacConstructorCall(digestArg) and
// we only want to consider it as an cryptographic operation if the input is available // we only want to consider it as an cryptographic operation if the input is available
exists(this.getParameter(1, "msg").asSink()) exists(this.(API::CallNode).getParameter(1, "msg").asSink())
} }
override DataFlow::Node getInitialization() { result = this } override DataFlow::Node getInitialization() { result = this }
override API::Node getDigestArg() { result = digestArg } override API::Node getDigestArg() { result = digestArg }
override DataFlow::Node getAnInput() { result = this.getParameter(1, "msg").asSink() } override DataFlow::Node getAnInput() {
result = this.(API::CallNode).getParameter(1, "msg").asSink()
}
} }
/** /**
@@ -2965,7 +2973,9 @@ module StdlibPrivate {
override API::Node getDigestArg() { result = digestArg } override API::Node getDigestArg() { result = digestArg }
override DataFlow::Node getAnInput() { result = this.getParameter(0, "msg").asSink() } override DataFlow::Node getAnInput() {
result = this.(API::CallNode).getParameter(0, "msg").asSink()
}
} }
/** /**
@@ -2978,9 +2988,11 @@ module StdlibPrivate {
override DataFlow::Node getInitialization() { result = this } override DataFlow::Node getInitialization() { result = this }
override API::Node getDigestArg() { result = this.getParameter(2, "digest") } override API::Node getDigestArg() { result = this.(API::CallNode).getParameter(2, "digest") }
override DataFlow::Node getAnInput() { result = this.getParameter(1, "msg").asSink() } override DataFlow::Node getAnInput() {
result = this.(API::CallNode).getParameter(1, "msg").asSink()
}
} }
// --------------------------------------------------------------------------- // ---------------------------------------------------------------------------

12
python/ql/lib/semmle/python/frameworks/Stdlib/Urllib.qll
View File

@@ -31,12 +31,14 @@ private module Urllib {
* See * See
* - https://docs.python.org/3.9/library/urllib.request.html#urllib.request.Request * - https://docs.python.org/3.9/library/urllib.request.html#urllib.request.Request
*/ */
private class RequestCall extends Http::Client::Request::Range, DataFlow::CallCfgNode { private class RequestCall extends Http::Client::Request::Range instanceof DataFlow::CallCfgNode {
RequestCall() { RequestCall() {
this = API::moduleImport("urllib").getMember("request").getMember("Request").getACall() this = API::moduleImport("urllib").getMember("request").getMember("Request").getACall()
} }
override DataFlow::Node getAUrlPart() { result in [this.getArg(0), this.getArgByName("url")] } override DataFlow::Node getAUrlPart() {
result in [super.getArg(0), super.getArgByName("url")]
}
override string getFramework() { result = "urllib.request.Request" } override string getFramework() { result = "urllib.request.Request" }
@@ -53,12 +55,14 @@ private module Urllib {
* See * See
* - https://docs.python.org/3.9/library/urllib.request.html#urllib.request.urlopen * - https://docs.python.org/3.9/library/urllib.request.html#urllib.request.urlopen
*/ */
private class UrlOpenCall extends Http::Client::Request::Range, DataFlow::CallCfgNode { private class UrlOpenCall extends Http::Client::Request::Range instanceof DataFlow::CallCfgNode {
UrlOpenCall() { UrlOpenCall() {
this = API::moduleImport("urllib").getMember("request").getMember("urlopen").getACall() this = API::moduleImport("urllib").getMember("request").getMember("urlopen").getACall()
} }
override DataFlow::Node getAUrlPart() { result in [this.getArg(0), this.getArgByName("url")] } override DataFlow::Node getAUrlPart() {
result in [super.getArg(0), super.getArgByName("url")]
}
override string getFramework() { result = "urllib.request.urlopen" } override string getFramework() { result = "urllib.request.urlopen" }

8
python/ql/lib/semmle/python/frameworks/Stdlib/Urllib2.qll
View File

@@ -20,10 +20,10 @@ private module Urllib2 {
* See * See
* - https://docs.python.org/2/library/urllib2.html#urllib2.Request * - https://docs.python.org/2/library/urllib2.html#urllib2.Request
*/ */
private class RequestCall extends Http::Client::Request::Range, DataFlow::CallCfgNode { private class RequestCall extends Http::Client::Request::Range instanceof DataFlow::CallCfgNode {
RequestCall() { this = API::moduleImport("urllib2").getMember("Request").getACall() } RequestCall() { this = API::moduleImport("urllib2").getMember("Request").getACall() }
override DataFlow::Node getAUrlPart() { result in [this.getArg(0), this.getArgByName("url")] } override DataFlow::Node getAUrlPart() { result in [super.getArg(0), super.getArgByName("url")] }
override string getFramework() { result = "urllib2.Request" } override string getFramework() { result = "urllib2.Request" }
@@ -40,10 +40,10 @@ private module Urllib2 {
* See * See
* - https://docs.python.org/2/library/urllib2.html#urllib2.urlopen * - https://docs.python.org/2/library/urllib2.html#urllib2.urlopen
*/ */
private class UrlOpenCall extends Http::Client::Request::Range, DataFlow::CallCfgNode { private class UrlOpenCall extends Http::Client::Request::Range instanceof DataFlow::CallCfgNode {
UrlOpenCall() { this = API::moduleImport("urllib2").getMember("urlopen").getACall() } UrlOpenCall() { this = API::moduleImport("urllib2").getMember("urlopen").getACall() }
override DataFlow::Node getAUrlPart() { result in [this.getArg(0), this.getArgByName("url")] } override DataFlow::Node getAUrlPart() { result in [super.getArg(0), super.getArgByName("url")] }
override string getFramework() { result = "urllib2.urlopen" } override string getFramework() { result = "urllib2.urlopen" }

6
python/ql/lib/semmle/python/frameworks/Urllib3.qll
View File

@@ -54,7 +54,7 @@ module Urllib3 {
* - https://urllib3.readthedocs.io/en/stable/reference/urllib3.request.html#urllib3.request.RequestMethods * - https://urllib3.readthedocs.io/en/stable/reference/urllib3.request.html#urllib3.request.RequestMethods
* - https://urllib3.readthedocs.io/en/stable/reference/urllib3.connectionpool.html#urllib3.HTTPConnectionPool.urlopen * - https://urllib3.readthedocs.io/en/stable/reference/urllib3.connectionpool.html#urllib3.HTTPConnectionPool.urlopen
*/ */
private class RequestCall extends Http::Client::Request::Range, API::CallNode { private class RequestCall extends Http::Client::Request::Range instanceof API::CallNode {
RequestCall() { RequestCall() {
this = this =
classRef() classRef()
@@ -63,7 +63,9 @@ module Urllib3 {
.getACall() .getACall()
} }
override DataFlow::Node getAUrlPart() { result in [this.getArg(1), this.getArgByName("url")] } override DataFlow::Node getAUrlPart() {
result in [super.getArg(1), super.getArgByName("url")]
}
override string getFramework() { result = "urllib3.PoolManager" } override string getFramework() { result = "urllib3.PoolManager" }

7
python/ql/lib/semmle/python/internal/ConceptsImports.qll
View File

@@ -1,7 +0,0 @@
/**
* This file contains imports required for the Python version of `ConceptsShared.qll`.
* Since they are language-specific, they can't be placed directly in that file, as it is shared between languages.
*/
import semmle.python.dataflow.new.DataFlow
import semmle.python.concepts.CryptoAlgorithms as CryptoAlgorithms

181
python/ql/lib/semmle/python/internal/ConceptsShared.qll
View File

@@ -1,181 +0,0 @@
/**
* Provides Concepts which are shared across languages.
*
* Each language has a language specific `Concepts.qll` file that can import the
* shared concepts from this file. A language can either re-export the concept directly,
* or can add additional member-predicates that are needed for that language.
*
* Moving forward, `Concepts.qll` will be the staging ground for brand new concepts from
* each language, but we will maintain a discipline of moving those concepts to
* `ConceptsShared.qll` ASAP.
*/
private import ConceptsImports
/**
* Provides models for cryptographic concepts.
*
* Note: The `CryptographicAlgorithm` class currently doesn't take weak keys into
* consideration for the `isWeak` member predicate. So RSA is always considered
* secure, although using a low number of bits will actually make it insecure. We plan
* to improve our libraries in the future to more precisely capture this aspect.
*/
module Cryptography {
class CryptographicAlgorithm = CryptoAlgorithms::CryptographicAlgorithm;
class EncryptionAlgorithm = CryptoAlgorithms::EncryptionAlgorithm;
class HashingAlgorithm = CryptoAlgorithms::HashingAlgorithm;
class PasswordHashingAlgorithm = CryptoAlgorithms::PasswordHashingAlgorithm;
/**
* A data flow node that is an application of a cryptographic algorithm. For example,
* encryption, decryption, signature-validation.
*
* Extend this class to refine existing API models. If you want to model new APIs,
* extend `CryptographicOperation::Range` instead.
*/
class CryptographicOperation extends DataFlow::Node instanceof CryptographicOperation::Range {
/** Gets the algorithm used, if it matches a known `CryptographicAlgorithm`. */
CryptographicAlgorithm getAlgorithm() { result = super.getAlgorithm() }
/** Gets the data flow node where the cryptographic algorithm used in this operation is configured. */
DataFlow::Node getInitialization() { result = super.getInitialization() }
/** Gets an input the algorithm is used on, for example the plain text input to be encrypted. */
DataFlow::Node getAnInput() { result = super.getAnInput() }
/**
* Gets the block mode used to perform this cryptographic operation.
*
* This predicate is only expected to have a result if two conditions hold:
* 1. The operation is an encryption operation, i.e. the algorithm used is an `EncryptionAlgorithm`, and
* 2. The algorithm used is a block cipher (not a stream cipher).
*
* If either of these conditions do not hold, then this predicate should have no result.
*/
BlockMode getBlockMode() { result = super.getBlockMode() }
}
/** Provides classes for modeling new applications of a cryptographic algorithms. */
module CryptographicOperation {
/**
* A data flow node that is an application of a cryptographic algorithm. For example,
* encryption, decryption, signature-validation.
*
* Extend this class to model new APIs. If you want to refine existing API models,
* extend `CryptographicOperation` instead.
*/
abstract class Range extends DataFlow::Node {
/** Gets the data flow node where the cryptographic algorithm used in this operation is configured. */
abstract DataFlow::Node getInitialization();
/** Gets the algorithm used, if it matches a known `CryptographicAlgorithm`. */
abstract CryptographicAlgorithm getAlgorithm();
/** Gets an input the algorithm is used on, for example the plain text input to be encrypted. */
abstract DataFlow::Node getAnInput();
/**
* Gets the block mode used to perform this cryptographic operation.
*
* This predicate is only expected to have a result if two conditions hold:
* 1. The operation is an encryption operation, i.e. the algorithm used is an `EncryptionAlgorithm`, and
* 2. The algorithm used is a block cipher (not a stream cipher).
*
* If either of these conditions do not hold, then this predicate should have no result.
*/
abstract BlockMode getBlockMode();
}
}
/**
* A cryptographic block cipher mode of operation. This can be used to encrypt
* data of arbitrary length using a block encryption algorithm.
*/
class BlockMode extends string {
BlockMode() {
this =
[
"ECB", "CBC", "GCM", "CCM", "CFB", "OFB", "CTR", "OPENPGP",
"XTS", // https://csrc.nist.gov/publications/detail/sp/800-38e/final
"EAX" // https://en.wikipedia.org/wiki/EAX_mode
]
}
/** Holds if this block mode is considered to be insecure. */
predicate isWeak() { this = "ECB" }
/** Holds if the given string appears to match this block mode. */
bindingset[s]
predicate matchesString(string s) { s.toUpperCase().matches("%" + this + "%") }
}
}
/** Provides classes for modeling HTTP-related APIs. */
module Http {
/** Provides classes for modeling HTTP clients. */
module Client {
/**
* A data flow node that makes an outgoing HTTP request.
*
* Extend this class to refine existing API models. If you want to model new APIs,
* extend `Http::Client::Request::Range` instead.
*/
class Request extends DataFlow::Node instanceof Request::Range {
/**
* Gets a data flow node that contributes to the URL of the request.
* Depending on the framework, a request may have multiple nodes which contribute to the URL.
*/
DataFlow::Node getAUrlPart() { result = super.getAUrlPart() }
/** Gets a string that identifies the framework used for this request. */
string getFramework() { result = super.getFramework() }
/**
* Holds if this request is made using a mode that disables SSL/TLS
* certificate validation, where `disablingNode` represents the point at
* which the validation was disabled, and `argumentOrigin` represents the origin
* of the argument that disabled the validation (which could be the same node as
* `disablingNode`).
*/
predicate disablesCertificateValidation(
DataFlow::Node disablingNode, DataFlow::Node argumentOrigin
) {
super.disablesCertificateValidation(disablingNode, argumentOrigin)
}
}
/** Provides a class for modeling new HTTP requests. */
module Request {
/**
* A data flow node that makes an outgoing HTTP request.
*
* Extend this class to model new APIs. If you want to refine existing API models,
* extend `Http::Client::Request` instead.
*/
abstract class Range extends DataFlow::Node {
/**
* Gets a data flow node that contributes to the URL of the request.
* Depending on the framework, a request may have multiple nodes which contribute to the URL.
*/
abstract DataFlow::Node getAUrlPart();
/** Gets a string that identifies the framework used for this request. */
abstract string getFramework();
/**
* Holds if this request is made using a mode that disables SSL/TLS
* certificate validation, where `disablingNode` represents the point at
* which the validation was disabled, and `argumentOrigin` represents the origin
* of the argument that disabled the validation (which could be the same node as
* `disablingNode`).
*/
abstract predicate disablesCertificateValidation(
DataFlow::Node disablingNode, DataFlow::Node argumentOrigin
);
}
}
}
}

188
python/ql/lib/semmle/python/security/internal/SensitiveDataHeuristics.qll
View File

@@ -1,188 +0,0 @@
/**
* INTERNAL: Do not use.
*
* Provides classes and predicates for identifying strings that may indicate the presence of sensitive data.
* Such that we can share this logic across our CodeQL analysis of different languages.
*
* 'Sensitive' data in general is anything that should not be sent around in unencrypted form.
*/
/**
* A classification of different kinds of sensitive data:
*
* - secret: generic secret or trusted data;
* - id: a user name or other account information;
* - password: a password or authorization key;
* - certificate: a certificate.
* - private: private data such as credit card numbers
*
* While classifications are represented as strings, this should not be relied upon.
* Instead, use the predicates in `SensitiveDataClassification::` to work with
* classifications.
*/
class SensitiveDataClassification extends string {
SensitiveDataClassification() { this in ["secret", "id", "password", "certificate", "private"] }
}
/**
* Provides predicates to select the different kinds of sensitive data we support.
*/
module SensitiveDataClassification {
/** Gets the classification for secret or trusted data. */
SensitiveDataClassification secret() { result = "secret" }
/** Gets the classification for user names or other account information. */
SensitiveDataClassification id() { result = "id" }
/** Gets the classification for passwords or authorization keys. */
SensitiveDataClassification password() { result = "password" }
/** Gets the classification for certificates. */
SensitiveDataClassification certificate() { result = "certificate" }
/** Gets the classification for private data. */
SensitiveDataClassification private() { result = "private" }
}
/**
* INTERNAL: Do not use.
*
* Provides heuristics for identifying names related to sensitive information.
*/
module HeuristicNames {
/**
* Gets a regular expression that identifies strings that may indicate the presence of secret
* or trusted data.
*/
string maybeSecret() { result = "(?is).*((?<!is|is_)secret|(?<!un|un_|is|is_)trusted).*" }
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* user names or other account information.
*/
string maybeAccountInfo() {
result = "(?is).*acc(ou)?nt.*" or
result = "(?is).*(puid|user.?name|user.?id|session.?(id|key)).*" or
result = "(?s).*([uU]|^|_|[a-z](?=U))([uU][iI][dD]).*"
}
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* a password or an authorization key.
*/
string maybePassword() {
result = "(?is).*pass(wd|word|code|.?phrase)(?!.*question).*" or
result = "(?is).*(auth(entication|ori[sz]ation)?).?key.*"
}
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* a certificate.
*/
string maybeCertificate() { result = "(?is).*(cert)(?!.*(format|name|ification)).*" }
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* private data.
*/
string maybePrivate() {
result =
"(?is).*(" +
// Inspired by the list on https://cwe.mitre.org/data/definitions/359.html
// Government identifiers, such as Social Security Numbers
"social.?security|employer.?identification|national.?insurance|resident.?id|" +
"passport.?(num|no)|([_-]|\\b)ssn([_-]|\\b)|" +
// Contact information, such as home addresses
"post.?code|zip.?code|home.?addr|" +
// and telephone numbers
"(mob(ile)?|home).?(num|no|tel|phone)|(tel|fax|phone).?(num|no)|telephone|" +
"emergency.?contact|" +
// Geographic location - where the user is (or was)
"latitude|longitude|nationality|" +
// Financial data - such as credit card numbers, salary, bank accounts, and debts
"(credit|debit|bank|visa).?(card|num|no|acc(ou)?nt)|acc(ou)?nt.?(no|num|credit)|" +
"salary|billing|credit.?(rating|score)|([_-]|\\b)ccn([_-]|\\b)|" +
// Communications - e-mail addresses, private e-mail messages, SMS text messages, chat logs, etc.
// "e(mail|_mail)|" + // this seems too noisy
// Health - medical conditions, insurance status, prescription records
"birth.?da(te|y)|da(te|y).?(of.?)?birth|" +
"medical|(health|care).?plan|healthkit|appointment|prescription|" +
"blood.?(type|alcohol|glucose|pressure)|heart.?(rate|rhythm)|body.?(mass|fat)|" +
"menstrua|pregnan|insulin|inhaler|" +
// Relationships - work and family
"employ(er|ee)|spouse|maiden.?name" +
// ---
").*"
}
/**
* Gets a regular expression that identifies strings that may indicate the presence
* of sensitive data, with `classification` describing the kind of sensitive data involved.
*/
string maybeSensitiveRegexp(SensitiveDataClassification classification) {
result = maybeSecret() and classification = SensitiveDataClassification::secret()
or
result = maybeAccountInfo() and classification = SensitiveDataClassification::id()
or
result = maybePassword() and classification = SensitiveDataClassification::password()
or
result = maybeCertificate() and
classification = SensitiveDataClassification::certificate()
or
result = maybePrivate() and
classification = SensitiveDataClassification::private()
}
/**
* Gets a regular expression that identifies strings that may indicate the presence of data
* that is hashed or encrypted, and hence rendered non-sensitive, or contains special characters
* suggesting nouns within the string do not represent the meaning of the whole string (e.g. a URL or a SQL query).
*
* We also filter out common words like `certain` and `concert`, since otherwise these could
* be matched by the certificate regular expressions. Same for `accountable` (account), or
* `secretarial` (secret).
*/
string notSensitiveRegexp() {
result =
"(?is).*([^\\w$.-]|redact|censor|obfuscate|hash|md5|sha|random|((?<!un)(en))?(crypt|(?<!pass)code)|certain|concert|secretar|accountant|accountab).*"
}
/**
* Holds if `name` may indicate the presence of sensitive data, and `name` does not indicate that
* the data is in fact non-sensitive (for example since it is hashed or encrypted).
*
* That is, one of the regexps from `maybeSensitiveRegexp` matches `name` (with the given
* classification), and none of the regexps from `notSensitiveRegexp` matches `name`.
*/
bindingset[name]
predicate nameIndicatesSensitiveData(string name) {
exists(string combinedRegexp |
// Combine all the maybe-sensitive regexps into one using non-capturing groups and |.
combinedRegexp =
"(?:" + strictconcat(string r | r = maybeSensitiveRegexp(_) | r, ")|(?:") + ")"
|
name.regexpMatch(combinedRegexp)
) and
not name.regexpMatch(notSensitiveRegexp())
}
/**
* Holds if `name` may indicate the presence of sensitive data, and
* `name` does not indicate that the data is in fact non-sensitive (for example since
* it is hashed or encrypted). `classification` describes the kind of sensitive data
* involved.
*
* That is, one of the regexps from `maybeSensitiveRegexp` matches `name` (with the
* given classification), and none of the regexps from `notSensitiveRegexp` matches
* `name`.
*
* When the set of names is large, it's worth using `nameIndicatesSensitiveData/1` as a first
* pass, since that combines all the regexps into one, and should be faster. Then call this
* predicate to get the classification(s).
*/
bindingset[name]
predicate nameIndicatesSensitiveData(string name, SensitiveDataClassification classification) {
name.regexpMatch(maybeSensitiveRegexp(classification)) and
not name.regexpMatch(notSensitiveRegexp())
}
}

8
ruby/ql/lib/codeql/ruby/Concepts.qll
View File

@@ -7,10 +7,14 @@
private import codeql.ruby.AST private import codeql.ruby.AST
private import codeql.ruby.CFG private import codeql.ruby.CFG
private import codeql.ruby.DataFlow private import codeql.ruby.DataFlow
private import codeql.ruby.dataflow.internal.DataFlowImplSpecific
private import codeql.ruby.Frameworks private import codeql.ruby.Frameworks
private import codeql.ruby.dataflow.RemoteFlowSources private import codeql.ruby.dataflow.RemoteFlowSources
private import codeql.ruby.ApiGraphs private import codeql.ruby.ApiGraphs
private import codeql.ruby.Regexp as RE private import codeql.ruby.Regexp as RE
private import codeql.concepts.ConceptsShared
private module ConceptsShared = ConceptsMake<Location, RubyDataFlow>;
/** /**
* A data-flow node that constructs a SQL statement. * A data-flow node that constructs a SQL statement.
@@ -682,7 +686,7 @@ module Http {
/** Provides classes for modeling HTTP clients. */ /** Provides classes for modeling HTTP clients. */
module Client { module Client {
import codeql.ruby.internal.ConceptsShared::Http::Client as SC import ConceptsShared::Http::Client as SC
/** /**
* A method call that makes an outgoing HTTP request. * A method call that makes an outgoing HTTP request.
@@ -1041,7 +1045,7 @@ module Cryptography {
// modify that part of the shared concept... which means we have to explicitly // modify that part of the shared concept... which means we have to explicitly
// re-export everything else. // re-export everything else.
// Using SC shorthand for "Shared Cryptography" // Using SC shorthand for "Shared Cryptography"
import codeql.ruby.internal.ConceptsShared::Cryptography as SC import ConceptsShared::Cryptography as SC
class CryptographicAlgorithm = SC::CryptographicAlgorithm; class CryptographicAlgorithm = SC::CryptographicAlgorithm;

14
ruby/ql/lib/codeql/ruby/frameworks/ActiveResource.qll
View File

@@ -183,8 +183,7 @@ module ActiveResource {
CollectionSource getCollection() { result = collection } CollectionSource getCollection() { result = collection }
} }
private class ModelClassMethodCallAsHttpRequest extends Http::Client::Request::Range, private class ModelClassMethodCallAsHttpRequest extends Http::Client::Request::Range instanceof ModelClassMethodCall
ModelClassMethodCall
{ {
ModelClassMethodCallAsHttpRequest() { ModelClassMethodCallAsHttpRequest() {
this.getMethodName() = ["all", "build", "create", "create!", "find", "first", "last"] this.getMethodName() = ["all", "build", "create", "create!", "find", "first", "last"]
@@ -195,20 +194,19 @@ module ActiveResource {
override predicate disablesCertificateValidation( override predicate disablesCertificateValidation(
DataFlow::Node disablingNode, DataFlow::Node argumentOrigin DataFlow::Node disablingNode, DataFlow::Node argumentOrigin
) { ) {
this.getModelClass().disablesCertificateValidation(disablingNode) and super.getModelClass().disablesCertificateValidation(disablingNode) and
// TODO: highlight real argument origin // TODO: highlight real argument origin
argumentOrigin = disablingNode argumentOrigin = disablingNode
} }
override DataFlow::Node getAUrlPart() { override DataFlow::Node getAUrlPart() {
result = this.getModelClass().getASiteAssignment().getAUrlPart() result = super.getModelClass().getASiteAssignment().getAUrlPart()
} }
override DataFlow::Node getResponseBody() { result = this } override DataFlow::Node getResponseBody() { result = this }
} }
private class ModelInstanceMethodCallAsHttpRequest extends Http::Client::Request::Range, private class ModelInstanceMethodCallAsHttpRequest extends Http::Client::Request::Range instanceof ModelInstanceMethodCall
ModelInstanceMethodCall
{ {
ModelInstanceMethodCallAsHttpRequest() { ModelInstanceMethodCallAsHttpRequest() {
this.getMethodName() = this.getMethodName() =
@@ -223,13 +221,13 @@ module ActiveResource {
override predicate disablesCertificateValidation( override predicate disablesCertificateValidation(
DataFlow::Node disablingNode, DataFlow::Node argumentOrigin DataFlow::Node disablingNode, DataFlow::Node argumentOrigin
) { ) {
this.getModelClass().disablesCertificateValidation(disablingNode) and super.getModelClass().disablesCertificateValidation(disablingNode) and
// TODO: highlight real argument origin // TODO: highlight real argument origin
argumentOrigin = disablingNode argumentOrigin = disablingNode
} }
override DataFlow::Node getAUrlPart() { override DataFlow::Node getAUrlPart() {
result = this.getModelClass().getASiteAssignment().getAUrlPart() result = super.getModelClass().getASiteAssignment().getAUrlPart()
} }
override DataFlow::Node getResponseBody() { result = this } override DataFlow::Node getResponseBody() { result = this }

6
ruby/ql/lib/codeql/ruby/frameworks/http_clients/Excon.qll
View File

@@ -23,7 +23,7 @@ private import codeql.ruby.DataFlow
* TODO: pipelining, streaming responses * TODO: pipelining, streaming responses
* https://github.com/excon/excon/blob/master/README.md * https://github.com/excon/excon/blob/master/README.md
*/ */
class ExconHttpRequest extends Http::Client::Request::Range, DataFlow::CallNode { class ExconHttpRequest extends Http::Client::Request::Range instanceof DataFlow::CallNode {
API::Node requestNode; API::Node requestNode;
API::Node connectionNode; API::Node connectionNode;
DataFlow::Node connectionUse; DataFlow::Node connectionUse;
@@ -54,9 +54,9 @@ class ExconHttpRequest extends Http::Client::Request::Range, DataFlow::CallNode
// For one-off requests, the URL is in the first argument of the request method call. // For one-off requests, the URL is in the first argument of the request method call.
// For connection re-use, the URL is split between the first argument of the `new` call // For connection re-use, the URL is split between the first argument of the `new` call
// and the `path` keyword argument of the request method call. // and the `path` keyword argument of the request method call.
result = this.getArgument(0) and not result.asExpr().getExpr() instanceof Pair result = super.getArgument(0) and not result.asExpr().getExpr() instanceof Pair
or or
result = this.getKeywordArgument("path") result = super.getKeywordArgument("path")
or or
result = connectionUse.(DataFlow::CallNode).getArgument(0) result = connectionUse.(DataFlow::CallNode).getArgument(0)
} }

4
ruby/ql/lib/codeql/ruby/frameworks/http_clients/Faraday.qll
View File

@@ -22,7 +22,7 @@ private import codeql.ruby.DataFlow
* connection.get("/").body * connection.get("/").body
* ``` * ```
*/ */
class FaradayHttpRequest extends Http::Client::Request::Range, DataFlow::CallNode { class FaradayHttpRequest extends Http::Client::Request::Range instanceof DataFlow::CallNode {
API::Node requestNode; API::Node requestNode;
API::Node connectionNode; API::Node connectionNode;
DataFlow::Node connectionUse; DataFlow::Node connectionUse;
@@ -47,7 +47,7 @@ class FaradayHttpRequest extends Http::Client::Request::Range, DataFlow::CallNod
override DataFlow::Node getResponseBody() { result = requestNode.getAMethodCall("body") } override DataFlow::Node getResponseBody() { result = requestNode.getAMethodCall("body") }
override DataFlow::Node getAUrlPart() { override DataFlow::Node getAUrlPart() {
result = this.getArgument(0) or result = super.getArgument(0) or
result = connectionUse.(DataFlow::CallNode).getArgument(0) or result = connectionUse.(DataFlow::CallNode).getArgument(0) or
result = connectionUse.(DataFlow::CallNode).getKeywordArgument("url") result = connectionUse.(DataFlow::CallNode).getKeywordArgument("url")
} }

4
ruby/ql/lib/codeql/ruby/frameworks/http_clients/HttpClient.qll
View File

@@ -14,7 +14,7 @@ private import codeql.ruby.DataFlow
* HTTPClient.get_content("http://example.com") * HTTPClient.get_content("http://example.com")
* ``` * ```
*/ */
class HttpClientRequest extends Http::Client::Request::Range, DataFlow::CallNode { class HttpClientRequest extends Http::Client::Request::Range instanceof DataFlow::CallNode {
API::Node requestNode; API::Node requestNode;
API::Node connectionNode; API::Node connectionNode;
string method; string method;
@@ -34,7 +34,7 @@ class HttpClientRequest extends Http::Client::Request::Range, DataFlow::CallNode
] ]
} }
override DataFlow::Node getAUrlPart() { result = this.getArgument(0) } override DataFlow::Node getAUrlPart() { result = super.getArgument(0) }
override DataFlow::Node getResponseBody() { override DataFlow::Node getResponseBody() {
// The `get_content` and `post_content` methods return the response body as // The `get_content` and `post_content` methods return the response body as

6
ruby/ql/lib/codeql/ruby/frameworks/http_clients/Httparty.qll
View File

@@ -23,7 +23,7 @@ private import codeql.ruby.DataFlow
* MyClass.new("http://example.com") * MyClass.new("http://example.com")
* ``` * ```
*/ */
class HttpartyRequest extends Http::Client::Request::Range, DataFlow::CallNode { class HttpartyRequest extends Http::Client::Request::Range instanceof DataFlow::CallNode {
API::Node requestNode; API::Node requestNode;
HttpartyRequest() { HttpartyRequest() {
@@ -33,7 +33,7 @@ class HttpartyRequest extends Http::Client::Request::Range, DataFlow::CallNode {
.getReturn(["get", "head", "delete", "options", "post", "put", "patch"]) .getReturn(["get", "head", "delete", "options", "post", "put", "patch"])
} }
override DataFlow::Node getAUrlPart() { result = this.getArgument(0) } override DataFlow::Node getAUrlPart() { result = super.getArgument(0) }
override DataFlow::Node getResponseBody() { override DataFlow::Node getResponseBody() {
// If HTTParty can recognise the response type, it will parse and return it // If HTTParty can recognise the response type, it will parse and return it
@@ -49,7 +49,7 @@ class HttpartyRequest extends Http::Client::Request::Range, DataFlow::CallNode {
/** Gets the value that controls certificate validation, if any. */ /** Gets the value that controls certificate validation, if any. */
DataFlow::Node getCertificateValidationControllingValue() { DataFlow::Node getCertificateValidationControllingValue() {
result = this.getKeywordArgumentIncludeHashArgument(["verify", "verify_peer"]) result = super.getKeywordArgumentIncludeHashArgument(["verify", "verify_peer"])
} }
cached cached

2
ruby/ql/lib/codeql/ruby/frameworks/http_clients/NetHttp.qll
View File

@@ -18,7 +18,7 @@ private import codeql.ruby.DataFlow
* response = req.get("/") * response = req.get("/")
* ``` * ```
*/ */
class NetHttpRequest extends Http::Client::Request::Range, DataFlow::CallNode { class NetHttpRequest extends Http::Client::Request::Range instanceof DataFlow::CallNode {
private DataFlow::CallNode request; private DataFlow::CallNode request;
private API::Node requestNode; private API::Node requestNode;
private boolean returnsResponseBody; private boolean returnsResponseBody;

15
ruby/ql/lib/codeql/ruby/frameworks/http_clients/OpenURI.qll
View File

@@ -18,7 +18,7 @@ private import codeql.ruby.frameworks.Core
* URI.parse("http://example.com").open.read * URI.parse("http://example.com").open.read
* ``` * ```
*/ */
class OpenUriRequest extends Http::Client::Request::Range, DataFlow::CallNode { class OpenUriRequest extends Http::Client::Request::Range instanceof DataFlow::CallNode {
API::Node requestNode; API::Node requestNode;
OpenUriRequest() { OpenUriRequest() {
@@ -30,7 +30,7 @@ class OpenUriRequest extends Http::Client::Request::Range, DataFlow::CallNode {
this = requestNode.asSource() this = requestNode.asSource()
} }
override DataFlow::Node getAUrlPart() { result = this.getArgument(0) } override DataFlow::Node getAUrlPart() { result = super.getArgument(0) }
override DataFlow::Node getResponseBody() { override DataFlow::Node getResponseBody() {
result = requestNode.getAMethodCall(["read", "readlines"]) result = requestNode.getAMethodCall(["read", "readlines"])
@@ -38,7 +38,7 @@ class OpenUriRequest extends Http::Client::Request::Range, DataFlow::CallNode {
/** Gets the value that controls certificate validation, if any. */ /** Gets the value that controls certificate validation, if any. */
DataFlow::Node getCertificateValidationControllingValue() { DataFlow::Node getCertificateValidationControllingValue() {
result = this.getKeywordArgumentIncludeHashArgument("ssl_verify_mode") result = super.getKeywordArgumentIncludeHashArgument("ssl_verify_mode")
} }
cached cached
@@ -60,11 +60,10 @@ class OpenUriRequest extends Http::Client::Request::Range, DataFlow::CallNode {
* Kernel.open("http://example.com").read * Kernel.open("http://example.com").read
* ``` * ```
*/ */
class OpenUriKernelOpenRequest extends Http::Client::Request::Range, DataFlow::CallNode instanceof KernelMethodCall class OpenUriKernelOpenRequest extends Http::Client::Request::Range instanceof KernelMethodCall {
{
OpenUriKernelOpenRequest() { this.getMethodName() = "open" } OpenUriKernelOpenRequest() { this.getMethodName() = "open" }
override DataFlow::Node getAUrlPart() { result = this.getArgument(0) } override DataFlow::Node getAUrlPart() { result = super.getArgument(0) }
override DataFlow::CallNode getResponseBody() { override DataFlow::CallNode getResponseBody() {
result.asExpr().getExpr().(MethodCall).getMethodName() in ["read", "readlines"] and result.asExpr().getExpr().(MethodCall).getMethodName() in ["read", "readlines"] and
@@ -73,14 +72,14 @@ class OpenUriKernelOpenRequest extends Http::Client::Request::Range, DataFlow::C
/** Gets the value that controls certificate validation, if any. */ /** Gets the value that controls certificate validation, if any. */
DataFlow::Node getCertificateValidationControllingValue() { DataFlow::Node getCertificateValidationControllingValue() {
result = this.getKeywordArgument("ssl_verify_mode") result = super.getKeywordArgument("ssl_verify_mode")
or or
// using a hashliteral // using a hashliteral
exists( exists(
DataFlow::LocalSourceNode optionsNode, CfgNodes::ExprNodes::PairCfgNode p, DataFlow::Node key DataFlow::LocalSourceNode optionsNode, CfgNodes::ExprNodes::PairCfgNode p, DataFlow::Node key
| |
// can't flow to argument 0, since that's the URL // can't flow to argument 0, since that's the URL
optionsNode.flowsTo(this.getArgument(any(int i | i > 0))) and optionsNode.flowsTo(super.getArgument(any(int i | i > 0))) and
p = optionsNode.asExpr().(CfgNodes::ExprNodes::HashLiteralCfgNode).getAKeyValuePair() and p = optionsNode.asExpr().(CfgNodes::ExprNodes::HashLiteralCfgNode).getAKeyValuePair() and
key.asExpr() = p.getKey() and key.asExpr() = p.getKey() and
key.getALocalSource().asExpr().getConstantValue().isStringlikeValue("ssl_verify_mode") and key.getALocalSource().asExpr().getConstantValue().isStringlikeValue("ssl_verify_mode") and

6
ruby/ql/lib/codeql/ruby/frameworks/http_clients/RestClient.qll
View File

@@ -16,7 +16,7 @@ private import codeql.ruby.DataFlow
* RestClient::Request.execute(url: "http://example.com").body * RestClient::Request.execute(url: "http://example.com").body
* ``` * ```
*/ */
class RestClientHttpRequest extends Http::Client::Request::Range, DataFlow::CallNode { class RestClientHttpRequest extends Http::Client::Request::Range instanceof DataFlow::CallNode {
API::Node requestNode; API::Node requestNode;
API::Node connectionNode; API::Node connectionNode;
@@ -37,9 +37,9 @@ class RestClientHttpRequest extends Http::Client::Request::Range, DataFlow::Call
} }
override DataFlow::Node getAUrlPart() { override DataFlow::Node getAUrlPart() {
result = this.getKeywordArgument("url") result = super.getKeywordArgument("url")
or or
result = this.getArgument(0) and result = super.getArgument(0) and
// this rules out the alternative above // this rules out the alternative above
not result.asExpr().getExpr() instanceof Pair not result.asExpr().getExpr() instanceof Pair
} }

6
ruby/ql/lib/codeql/ruby/frameworks/http_clients/Typhoeus.qll
View File

@@ -14,7 +14,7 @@ private import codeql.ruby.DataFlow
* Typhoeus.get("http://example.com").body * Typhoeus.get("http://example.com").body
* ``` * ```
*/ */
class TyphoeusHttpRequest extends Http::Client::Request::Range, DataFlow::CallNode { class TyphoeusHttpRequest extends Http::Client::Request::Range instanceof DataFlow::CallNode {
API::Node requestNode; API::Node requestNode;
boolean directResponse; boolean directResponse;
@@ -31,7 +31,7 @@ class TyphoeusHttpRequest extends Http::Client::Request::Range, DataFlow::CallNo
) )
} }
override DataFlow::Node getAUrlPart() { result = this.getArgument(0) } override DataFlow::Node getAUrlPart() { result = super.getArgument(0) }
override DataFlow::Node getResponseBody() { override DataFlow::Node getResponseBody() {
directResponse = true and directResponse = true and
@@ -43,7 +43,7 @@ class TyphoeusHttpRequest extends Http::Client::Request::Range, DataFlow::CallNo
/** Gets the value that controls certificate validation, if any. */ /** Gets the value that controls certificate validation, if any. */
DataFlow::Node getCertificateValidationControllingValue() { DataFlow::Node getCertificateValidationControllingValue() {
result = this.getKeywordArgumentIncludeHashArgument("ssl_verifypeer") result = super.getKeywordArgumentIncludeHashArgument("ssl_verifypeer")
} }
cached cached

7
ruby/ql/lib/codeql/ruby/internal/ConceptsImports.qll
View File

@@ -1,7 +0,0 @@
/**
* This file contains imports required for the Ruby version of `ConceptsShared.qll`.
* Since they are language-specific, they can't be placed directly in that file, as it is shared between languages.
*/
import codeql.ruby.DataFlow
import codeql.ruby.security.CryptoAlgorithms as CryptoAlgorithms

181
ruby/ql/lib/codeql/ruby/internal/ConceptsShared.qll
View File

@@ -1,181 +0,0 @@
/**
* Provides Concepts which are shared across languages.
*
* Each language has a language specific `Concepts.qll` file that can import the
* shared concepts from this file. A language can either re-export the concept directly,
* or can add additional member-predicates that are needed for that language.
*
* Moving forward, `Concepts.qll` will be the staging ground for brand new concepts from
* each language, but we will maintain a discipline of moving those concepts to
* `ConceptsShared.qll` ASAP.
*/
private import ConceptsImports
/**
* Provides models for cryptographic concepts.
*
* Note: The `CryptographicAlgorithm` class currently doesn't take weak keys into
* consideration for the `isWeak` member predicate. So RSA is always considered
* secure, although using a low number of bits will actually make it insecure. We plan
* to improve our libraries in the future to more precisely capture this aspect.
*/
module Cryptography {
class CryptographicAlgorithm = CryptoAlgorithms::CryptographicAlgorithm;
class EncryptionAlgorithm = CryptoAlgorithms::EncryptionAlgorithm;
class HashingAlgorithm = CryptoAlgorithms::HashingAlgorithm;
class PasswordHashingAlgorithm = CryptoAlgorithms::PasswordHashingAlgorithm;
/**
* A data flow node that is an application of a cryptographic algorithm. For example,
* encryption, decryption, signature-validation.
*
* Extend this class to refine existing API models. If you want to model new APIs,
* extend `CryptographicOperation::Range` instead.
*/
class CryptographicOperation extends DataFlow::Node instanceof CryptographicOperation::Range {
/** Gets the algorithm used, if it matches a known `CryptographicAlgorithm`. */
CryptographicAlgorithm getAlgorithm() { result = super.getAlgorithm() }
/** Gets the data flow node where the cryptographic algorithm used in this operation is configured. */
DataFlow::Node getInitialization() { result = super.getInitialization() }
/** Gets an input the algorithm is used on, for example the plain text input to be encrypted. */
DataFlow::Node getAnInput() { result = super.getAnInput() }
/**
* Gets the block mode used to perform this cryptographic operation.
*
* This predicate is only expected to have a result if two conditions hold:
* 1. The operation is an encryption operation, i.e. the algorithm used is an `EncryptionAlgorithm`, and
* 2. The algorithm used is a block cipher (not a stream cipher).
*
* If either of these conditions do not hold, then this predicate should have no result.
*/
BlockMode getBlockMode() { result = super.getBlockMode() }
}
/** Provides classes for modeling new applications of a cryptographic algorithms. */
module CryptographicOperation {
/**
* A data flow node that is an application of a cryptographic algorithm. For example,
* encryption, decryption, signature-validation.
*
* Extend this class to model new APIs. If you want to refine existing API models,
* extend `CryptographicOperation` instead.
*/
abstract class Range extends DataFlow::Node {
/** Gets the data flow node where the cryptographic algorithm used in this operation is configured. */
abstract DataFlow::Node getInitialization();
/** Gets the algorithm used, if it matches a known `CryptographicAlgorithm`. */
abstract CryptographicAlgorithm getAlgorithm();
/** Gets an input the algorithm is used on, for example the plain text input to be encrypted. */
abstract DataFlow::Node getAnInput();
/**
* Gets the block mode used to perform this cryptographic operation.
*
* This predicate is only expected to have a result if two conditions hold:
* 1. The operation is an encryption operation, i.e. the algorithm used is an `EncryptionAlgorithm`, and
* 2. The algorithm used is a block cipher (not a stream cipher).
*
* If either of these conditions do not hold, then this predicate should have no result.
*/
abstract BlockMode getBlockMode();
}
}
/**
* A cryptographic block cipher mode of operation. This can be used to encrypt
* data of arbitrary length using a block encryption algorithm.
*/
class BlockMode extends string {
BlockMode() {
this =
[
"ECB", "CBC", "GCM", "CCM", "CFB", "OFB", "CTR", "OPENPGP",
"XTS", // https://csrc.nist.gov/publications/detail/sp/800-38e/final
"EAX" // https://en.wikipedia.org/wiki/EAX_mode
]
}
/** Holds if this block mode is considered to be insecure. */
predicate isWeak() { this = "ECB" }
/** Holds if the given string appears to match this block mode. */
bindingset[s]
predicate matchesString(string s) { s.toUpperCase().matches("%" + this + "%") }
}
}
/** Provides classes for modeling HTTP-related APIs. */
module Http {
/** Provides classes for modeling HTTP clients. */
module Client {
/**
* A data flow node that makes an outgoing HTTP request.
*
* Extend this class to refine existing API models. If you want to model new APIs,
* extend `Http::Client::Request::Range` instead.
*/
class Request extends DataFlow::Node instanceof Request::Range {
/**
* Gets a data flow node that contributes to the URL of the request.
* Depending on the framework, a request may have multiple nodes which contribute to the URL.
*/
DataFlow::Node getAUrlPart() { result = super.getAUrlPart() }
/** Gets a string that identifies the framework used for this request. */
string getFramework() { result = super.getFramework() }
/**
* Holds if this request is made using a mode that disables SSL/TLS
* certificate validation, where `disablingNode` represents the point at
* which the validation was disabled, and `argumentOrigin` represents the origin
* of the argument that disabled the validation (which could be the same node as
* `disablingNode`).
*/
predicate disablesCertificateValidation(
DataFlow::Node disablingNode, DataFlow::Node argumentOrigin
) {
super.disablesCertificateValidation(disablingNode, argumentOrigin)
}
}
/** Provides a class for modeling new HTTP requests. */
module Request {
/**
* A data flow node that makes an outgoing HTTP request.
*
* Extend this class to model new APIs. If you want to refine existing API models,
* extend `Http::Client::Request` instead.
*/
abstract class Range extends DataFlow::Node {
/**
* Gets a data flow node that contributes to the URL of the request.
* Depending on the framework, a request may have multiple nodes which contribute to the URL.
*/
abstract DataFlow::Node getAUrlPart();
/** Gets a string that identifies the framework used for this request. */
abstract string getFramework();
/**
* Holds if this request is made using a mode that disables SSL/TLS
* certificate validation, where `disablingNode` represents the point at
* which the validation was disabled, and `argumentOrigin` represents the origin
* of the argument that disabled the validation (which could be the same node as
* `disablingNode`).
*/
abstract predicate disablesCertificateValidation(
DataFlow::Node disablingNode, DataFlow::Node argumentOrigin
);
}
}
}
}

114
ruby/ql/lib/codeql/ruby/security/CryptoAlgorithms.qll
View File

@@ -1,117 +1,5 @@
/** /**
* Provides classes modeling cryptographic algorithms, separated into strong and weak variants. * Provides classes modeling cryptographic algorithms, separated into strong and weak variants.
*
* The classification into strong and weak are based on Wikipedia, OWASP and Google (2021).
*/ */
private import internal.CryptoAlgorithmNames import codeql.concepts.CryptoAlgorithms
/**
* A cryptographic algorithm.
*/
private newtype TCryptographicAlgorithm =
MkHashingAlgorithm(string name, boolean isWeak) {
isStrongHashingAlgorithm(name) and isWeak = false
or
isWeakHashingAlgorithm(name) and isWeak = true
} or
MkEncryptionAlgorithm(string name, boolean isWeak) {
isStrongEncryptionAlgorithm(name) and isWeak = false
or
isWeakEncryptionAlgorithm(name) and isWeak = true
} or
MkPasswordHashingAlgorithm(string name, boolean isWeak) {
isStrongPasswordHashingAlgorithm(name) and isWeak = false
or
isWeakPasswordHashingAlgorithm(name) and isWeak = true
}
/**
* Gets the most specific `CryptographicAlgorithm` that matches the given `name`.
* A matching algorithm is one where the name of the algorithm matches the start of name, with allowances made for different name formats.
* In the case that multiple `CryptographicAlgorithm`s match the given `name`, the algorithm(s) with the longest name will be selected. This is intended to select more specific versions of algorithms when multiple versions could match - for example "SHA3_224" matches against both "SHA3" and "SHA3224", but the latter is a more precise match.
*/
bindingset[name]
private CryptographicAlgorithm getBestAlgorithmForName(string name) {
result =
max(CryptographicAlgorithm algorithm |
algorithm.getName() =
[
name.toUpperCase(), // the full name
name.toUpperCase().regexpCapture("^([\\w]+)(?:-.*)?$", 1), // the name prior to any dashes or spaces
name.toUpperCase().regexpCapture("^([A-Z0-9]+)(?:(-|_).*)?$", 1) // the name prior to any dashes, spaces, or underscores
].regexpReplaceAll("[-_ ]", "") // strip dashes, underscores, and spaces
|
algorithm order by algorithm.getName().length()
)
}
/**
* A cryptographic algorithm.
*/
abstract class CryptographicAlgorithm extends TCryptographicAlgorithm {
/** Gets a textual representation of this element. */
string toString() { result = this.getName() }
/**
* Gets the normalized name of this algorithm (upper-case, no spaces, dashes or underscores).
*/
abstract string getName();
/**
* Holds if the name of this algorithm is the most specific match for `name`.
* This predicate matches quite liberally to account for different ways of formatting algorithm names, e.g. using dashes, underscores, or spaces as separators, including or not including block modes of operation, etc.
*/
bindingset[name]
predicate matchesName(string name) { this = getBestAlgorithmForName(name) }
/**
* Holds if this algorithm is weak.
*/
abstract predicate isWeak();
}
/**
* A hashing algorithm such as `MD5` or `SHA512`.
*/
class HashingAlgorithm extends MkHashingAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
HashingAlgorithm() { this = MkHashingAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
}
/**
* An encryption algorithm such as `DES` or `AES512`.
*/
class EncryptionAlgorithm extends MkEncryptionAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
EncryptionAlgorithm() { this = MkEncryptionAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
/** Holds if this algorithm is a stream cipher. */
predicate isStreamCipher() { isStreamCipher(name) }
}
/**
* A password hashing algorithm such as `PBKDF2` or `SCRYPT`.
*/
class PasswordHashingAlgorithm extends MkPasswordHashingAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
PasswordHashingAlgorithm() { this = MkPasswordHashingAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
}

9
ruby/ql/lib/codeql/ruby/security/OpenSSL.qll
View File

@@ -3,7 +3,7 @@
* an underlying OpenSSL or LibreSSL C library. * an underlying OpenSSL or LibreSSL C library.
*/ */
private import internal.CryptoAlgorithmNames private import codeql.concepts.internal.CryptoAlgorithmNames
private import codeql.ruby.Concepts private import codeql.ruby.Concepts
private import codeql.ruby.DataFlow private import codeql.ruby.DataFlow
private import codeql.ruby.ApiGraphs private import codeql.ruby.ApiGraphs
@@ -544,8 +544,7 @@ private class CipherNode extends DataFlow::Node {
} }
/** An operation using the OpenSSL library that uses a cipher. */ /** An operation using the OpenSSL library that uses a cipher. */
private class CipherOperation extends Cryptography::CryptographicOperation::Range, private class CipherOperation extends Cryptography::CryptographicOperation::Range instanceof DataFlow::CallNode
DataFlow::CallNode
{ {
private CipherNode cipherNode; private CipherNode cipherNode;
@@ -564,8 +563,8 @@ private class CipherOperation extends Cryptography::CryptographicOperation::Rang
} }
override DataFlow::Node getAnInput() { override DataFlow::Node getAnInput() {
this.getMethodName() = "update" and super.getMethodName() = "update" and
result = this.getArgument(0) result = super.getArgument(0)
} }
override Cryptography::BlockMode getBlockMode() { override Cryptography::BlockMode getBlockMode() {

2
ruby/ql/lib/codeql/ruby/security/SensitiveActions.qll
View File

@@ -11,7 +11,7 @@
private import codeql.ruby.AST private import codeql.ruby.AST
private import codeql.ruby.DataFlow private import codeql.ruby.DataFlow
import codeql.ruby.security.internal.SensitiveDataHeuristics import codeql.concepts.internal.SensitiveDataHeuristics
private import HeuristicNames private import HeuristicNames
private import codeql.ruby.CFG private import codeql.ruby.CFG

4
ruby/ql/lib/codeql/ruby/security/internal/CleartextSources.qll
View File

@@ -8,8 +8,8 @@ private import codeql.ruby.AST
private import codeql.ruby.DataFlow private import codeql.ruby.DataFlow
private import codeql.ruby.TaintTracking::TaintTracking private import codeql.ruby.TaintTracking::TaintTracking
private import codeql.ruby.dataflow.RemoteFlowSources private import codeql.ruby.dataflow.RemoteFlowSources
private import SensitiveDataHeuristics::HeuristicNames private import codeql.concepts.internal.SensitiveDataHeuristics::HeuristicNames
private import SensitiveDataHeuristics private import codeql.concepts.internal.SensitiveDataHeuristics
private import codeql.ruby.CFG private import codeql.ruby.CFG
private import codeql.ruby.dataflow.SSA private import codeql.ruby.dataflow.SSA

84
ruby/ql/lib/codeql/ruby/security/internal/CryptoAlgorithmNames.qll
View File

@@ -1,84 +0,0 @@
/**
* Names of cryptographic algorithms, separated into strong and weak variants.
*
* The names are normalized: upper-case, no spaces, dashes or underscores.
*
* The names are inspired by the names used in real world crypto libraries.
*
* The classification into strong and weak are based on Wikipedia, OWASP and Google (2021).
*/
/**
* Holds if `name` corresponds to a strong hashing algorithm.
*/
predicate isStrongHashingAlgorithm(string name) {
name =
[
// see https://cryptography.io/en/latest/hazmat/primitives/cryptographic-hashes/#blake2
// and https://www.blake2.net/
"BLAKE2", "BLAKE2B", "BLAKE2S",
// see https://github.com/BLAKE3-team/BLAKE3
"BLAKE3",
//
"DSA", "ED25519", "ES256", "ECDSA256", "ES384", "ECDSA384", "ES512", "ECDSA512", "SHA2",
"SHA224", "SHA256", "SHA384", "SHA512", "SHA3", "SHA3224", "SHA3256", "SHA3384", "SHA3512",
// see https://cryptography.io/en/latest/hazmat/primitives/cryptographic-hashes/#cryptography.hazmat.primitives.hashes.SHAKE128
"SHAKE128", "SHAKE256",
// see https://cryptography.io/en/latest/hazmat/primitives/cryptographic-hashes/#sm3
"SM3",
// see https://security.stackexchange.com/a/216297
"WHIRLPOOL",
]
}
/**
* Holds if `name` corresponds to a weak hashing algorithm.
*/
predicate isWeakHashingAlgorithm(string name) {
name =
[
"HAVEL128", "MD2", "MD4", "MD5", "PANAMA", "RIPEMD", "RIPEMD128", "RIPEMD256", "RIPEMD160",
"RIPEMD320", "SHA0", "SHA1"
]
}
/**
* Holds if `name` corresponds to a strong encryption algorithm.
*/
predicate isStrongEncryptionAlgorithm(string name) {
name =
[
"AES", "AES128", "AES192", "AES256", "AES512", "AES-128", "AES-192", "AES-256", "AES-512",
"ARIA", "BLOWFISH", "BF", "ECIES", "CAST", "CAST5", "CAMELLIA", "CAMELLIA128", "CAMELLIA192",
"CAMELLIA256", "CAMELLIA-128", "CAMELLIA-192", "CAMELLIA-256", "CHACHA", "GOST", "GOST89",
"IDEA", "RABBIT", "RSA", "SEED", "SM4"
]
}
/**
* Holds if `name` corresponds to a weak encryption algorithm.
*/
predicate isWeakEncryptionAlgorithm(string name) {
name =
[
"DES", "3DES", "DES3", "TRIPLEDES", "DESX", "TDEA", "TRIPLEDEA", "ARC2", "RC2", "ARC4", "RC4",
"ARCFOUR", "ARC5", "RC5"
]
}
/**
* Holds if `name` corresponds to a strong password hashing algorithm.
*/
predicate isStrongPasswordHashingAlgorithm(string name) {
name = ["ARGON2", "PBKDF2", "BCRYPT", "SCRYPT"]
}
/**
* Holds if `name` corresponds to a weak password hashing algorithm.
*/
predicate isWeakPasswordHashingAlgorithm(string name) { name = "EVPKDF" }
/**
* Holds if `name` corresponds to a stream cipher.
*/
predicate isStreamCipher(string name) { name = ["CHACHA", "RC4", "ARC4", "ARCFOUR", "RABBIT"] }

1
ruby/ql/lib/qlpack.yml
View File

@@ -6,6 +6,7 @@ dbscheme: ruby.dbscheme
upgrades: upgrades upgrades: upgrades
library: true library: true
dependencies: dependencies:
codeql/concepts: ${workspace}
codeql/controlflow: ${workspace} codeql/controlflow: ${workspace}
codeql/dataflow: ${workspace} codeql/dataflow: ${workspace}
codeql/mad: ${workspace} codeql/mad: ${workspace}

2
ruby/ql/test/library-tests/security/CryptoAlgorithms.ql
View File

@@ -1,5 +1,5 @@
import codeql.ruby.AST import codeql.ruby.AST
import codeql.ruby.security.CryptoAlgorithms import codeql.concepts.CryptoAlgorithms
query predicate weakHashingAlgorithms(HashingAlgorithm ha) { ha.isWeak() } query predicate weakHashingAlgorithms(HashingAlgorithm ha) { ha.isWeak() }

7
rust/ql/lib/codeql/rust/Concepts.qll
View File

@@ -5,11 +5,16 @@
*/ */
private import codeql.rust.dataflow.DataFlow private import codeql.rust.dataflow.DataFlow
private import codeql.rust.dataflow.internal.DataFlowImpl
private import codeql.Locations
private import codeql.threatmodels.ThreatModels private import codeql.threatmodels.ThreatModels
private import codeql.rust.Frameworks private import codeql.rust.Frameworks
private import codeql.rust.dataflow.FlowSource private import codeql.rust.dataflow.FlowSource
private import codeql.rust.controlflow.ControlFlowGraph as Cfg private import codeql.rust.controlflow.ControlFlowGraph as Cfg
private import codeql.rust.controlflow.CfgNodes as CfgNodes private import codeql.rust.controlflow.CfgNodes as CfgNodes
private import codeql.concepts.ConceptsShared
private module ConceptsShared = ConceptsMake<Location, RustDataFlow>;
/** /**
* A data flow source for a specific threat-model. * A data flow source for a specific threat-model.
@@ -302,7 +307,7 @@ module SqlSanitization {
* Provides models for cryptographic things. * Provides models for cryptographic things.
*/ */
module Cryptography { module Cryptography {
private import codeql.rust.internal.ConceptsShared::Cryptography as SC import ConceptsShared::Cryptography as SC
final class CryptographicOperation = SC::CryptographicOperation; final class CryptographicOperation = SC::CryptographicOperation;

7
rust/ql/lib/codeql/rust/internal/ConceptsImports.qll
View File

@@ -1,7 +0,0 @@
/**
* This file contains imports required for the Rust version of `ConceptsShared.qll`.
* Since they are language-specific, they can't be placed directly in that file, as it is shared between languages.
*/
import codeql.rust.dataflow.DataFlow::DataFlow as DataFlow
import codeql.rust.security.CryptoAlgorithms as CryptoAlgorithms

181
rust/ql/lib/codeql/rust/internal/ConceptsShared.qll
View File

@@ -1,181 +0,0 @@
/**
* Provides Concepts which are shared across languages.
*
* Each language has a language specific `Concepts.qll` file that can import the
* shared concepts from this file. A language can either re-export the concept directly,
* or can add additional member-predicates that are needed for that language.
*
* Moving forward, `Concepts.qll` will be the staging ground for brand new concepts from
* each language, but we will maintain a discipline of moving those concepts to
* `ConceptsShared.qll` ASAP.
*/
private import ConceptsImports
/**
* Provides models for cryptographic concepts.
*
* Note: The `CryptographicAlgorithm` class currently doesn't take weak keys into
* consideration for the `isWeak` member predicate. So RSA is always considered
* secure, although using a low number of bits will actually make it insecure. We plan
* to improve our libraries in the future to more precisely capture this aspect.
*/
module Cryptography {
class CryptographicAlgorithm = CryptoAlgorithms::CryptographicAlgorithm;
class EncryptionAlgorithm = CryptoAlgorithms::EncryptionAlgorithm;
class HashingAlgorithm = CryptoAlgorithms::HashingAlgorithm;
class PasswordHashingAlgorithm = CryptoAlgorithms::PasswordHashingAlgorithm;
/**
* A data flow node that is an application of a cryptographic algorithm. For example,
* encryption, decryption, signature-validation.
*
* Extend this class to refine existing API models. If you want to model new APIs,
* extend `CryptographicOperation::Range` instead.
*/
class CryptographicOperation extends DataFlow::Node instanceof CryptographicOperation::Range {
/** Gets the algorithm used, if it matches a known `CryptographicAlgorithm`. */
CryptographicAlgorithm getAlgorithm() { result = super.getAlgorithm() }
/** Gets the data flow node where the cryptographic algorithm used in this operation is configured. */
DataFlow::Node getInitialization() { result = super.getInitialization() }
/** Gets an input the algorithm is used on, for example the plain text input to be encrypted. */
DataFlow::Node getAnInput() { result = super.getAnInput() }
/**
* Gets the block mode used to perform this cryptographic operation.
*
* This predicate is only expected to have a result if two conditions hold:
* 1. The operation is an encryption operation, i.e. the algorithm used is an `EncryptionAlgorithm`, and
* 2. The algorithm used is a block cipher (not a stream cipher).
*
* If either of these conditions do not hold, then this predicate should have no result.
*/
BlockMode getBlockMode() { result = super.getBlockMode() }
}
/** Provides classes for modeling new applications of a cryptographic algorithms. */
module CryptographicOperation {
/**
* A data flow node that is an application of a cryptographic algorithm. For example,
* encryption, decryption, signature-validation.
*
* Extend this class to model new APIs. If you want to refine existing API models,
* extend `CryptographicOperation` instead.
*/
abstract class Range extends DataFlow::Node {
/** Gets the data flow node where the cryptographic algorithm used in this operation is configured. */
abstract DataFlow::Node getInitialization();
/** Gets the algorithm used, if it matches a known `CryptographicAlgorithm`. */
abstract CryptographicAlgorithm getAlgorithm();
/** Gets an input the algorithm is used on, for example the plain text input to be encrypted. */
abstract DataFlow::Node getAnInput();
/**
* Gets the block mode used to perform this cryptographic operation.
*
* This predicate is only expected to have a result if two conditions hold:
* 1. The operation is an encryption operation, i.e. the algorithm used is an `EncryptionAlgorithm`, and
* 2. The algorithm used is a block cipher (not a stream cipher).
*
* If either of these conditions do not hold, then this predicate should have no result.
*/
abstract BlockMode getBlockMode();
}
}
/**
* A cryptographic block cipher mode of operation. This can be used to encrypt
* data of arbitrary length using a block encryption algorithm.
*/
class BlockMode extends string {
BlockMode() {
this =
[
"ECB", "CBC", "GCM", "CCM", "CFB", "OFB", "CTR", "OPENPGP",
"XTS", // https://csrc.nist.gov/publications/detail/sp/800-38e/final
"EAX" // https://en.wikipedia.org/wiki/EAX_mode
]
}
/** Holds if this block mode is considered to be insecure. */
predicate isWeak() { this = "ECB" }
/** Holds if the given string appears to match this block mode. */
bindingset[s]
predicate matchesString(string s) { s.toUpperCase().matches("%" + this + "%") }
}
}
/** Provides classes for modeling HTTP-related APIs. */
module Http {
/** Provides classes for modeling HTTP clients. */
module Client {
/**
* A data flow node that makes an outgoing HTTP request.
*
* Extend this class to refine existing API models. If you want to model new APIs,
* extend `Http::Client::Request::Range` instead.
*/
class Request extends DataFlow::Node instanceof Request::Range {
/**
* Gets a data flow node that contributes to the URL of the request.
* Depending on the framework, a request may have multiple nodes which contribute to the URL.
*/
DataFlow::Node getAUrlPart() { result = super.getAUrlPart() }
/** Gets a string that identifies the framework used for this request. */
string getFramework() { result = super.getFramework() }
/**
* Holds if this request is made using a mode that disables SSL/TLS
* certificate validation, where `disablingNode` represents the point at
* which the validation was disabled, and `argumentOrigin` represents the origin
* of the argument that disabled the validation (which could be the same node as
* `disablingNode`).
*/
predicate disablesCertificateValidation(
DataFlow::Node disablingNode, DataFlow::Node argumentOrigin
) {
super.disablesCertificateValidation(disablingNode, argumentOrigin)
}
}
/** Provides a class for modeling new HTTP requests. */
module Request {
/**
* A data flow node that makes an outgoing HTTP request.
*
* Extend this class to model new APIs. If you want to refine existing API models,
* extend `Http::Client::Request` instead.
*/
abstract class Range extends DataFlow::Node {
/**
* Gets a data flow node that contributes to the URL of the request.
* Depending on the framework, a request may have multiple nodes which contribute to the URL.
*/
abstract DataFlow::Node getAUrlPart();
/** Gets a string that identifies the framework used for this request. */
abstract string getFramework();
/**
* Holds if this request is made using a mode that disables SSL/TLS
* certificate validation, where `disablingNode` represents the point at
* which the validation was disabled, and `argumentOrigin` represents the origin
* of the argument that disabled the validation (which could be the same node as
* `disablingNode`).
*/
abstract predicate disablesCertificateValidation(
DataFlow::Node disablingNode, DataFlow::Node argumentOrigin
);
}
}
}
}

114
rust/ql/lib/codeql/rust/security/CryptoAlgorithms.qll
View File

@@ -1,117 +1,5 @@
/** /**
* Provides classes modeling cryptographic algorithms, separated into strong and weak variants. * Provides classes modeling cryptographic algorithms, separated into strong and weak variants.
*
* The classification into strong and weak are based on Wikipedia, OWASP and Google (2021).
*/ */
private import internal.CryptoAlgorithmNames import codeql.concepts.CryptoAlgorithms
/**
* A cryptographic algorithm.
*/
private newtype TCryptographicAlgorithm =
MkHashingAlgorithm(string name, boolean isWeak) {
isStrongHashingAlgorithm(name) and isWeak = false
or
isWeakHashingAlgorithm(name) and isWeak = true
} or
MkEncryptionAlgorithm(string name, boolean isWeak) {
isStrongEncryptionAlgorithm(name) and isWeak = false
or
isWeakEncryptionAlgorithm(name) and isWeak = true
} or
MkPasswordHashingAlgorithm(string name, boolean isWeak) {
isStrongPasswordHashingAlgorithm(name) and isWeak = false
or
isWeakPasswordHashingAlgorithm(name) and isWeak = true
}
/**
* Gets the most specific `CryptographicAlgorithm` that matches the given `name`.
* A matching algorithm is one where the name of the algorithm matches the start of name, with allowances made for different name formats.
* In the case that multiple `CryptographicAlgorithm`s match the given `name`, the algorithm(s) with the longest name will be selected. This is intended to select more specific versions of algorithms when multiple versions could match - for example "SHA3_224" matches against both "SHA3" and "SHA3224", but the latter is a more precise match.
*/
bindingset[name]
private CryptographicAlgorithm getBestAlgorithmForName(string name) {
result =
max(CryptographicAlgorithm algorithm |
algorithm.getName() =
[
name.toUpperCase(), // the full name
name.toUpperCase().regexpCapture("^([\\w]+)(?:-.*)?$", 1), // the name prior to any dashes or spaces
name.toUpperCase().regexpCapture("^([A-Z0-9]+)(?:(-|_).*)?$", 1) // the name prior to any dashes, spaces, or underscores
].regexpReplaceAll("[-_ ]", "") // strip dashes, underscores, and spaces
|
algorithm order by algorithm.getName().length()
)
}
/**
* A cryptographic algorithm.
*/
abstract class CryptographicAlgorithm extends TCryptographicAlgorithm {
/** Gets a textual representation of this element. */
string toString() { result = this.getName() }
/**
* Gets the normalized name of this algorithm (upper-case, no spaces, dashes or underscores).
*/
abstract string getName();
/**
* Holds if the name of this algorithm is the most specific match for `name`.
* This predicate matches quite liberally to account for different ways of formatting algorithm names, e.g. using dashes, underscores, or spaces as separators, including or not including block modes of operation, etc.
*/
bindingset[name]
predicate matchesName(string name) { this = getBestAlgorithmForName(name) }
/**
* Holds if this algorithm is weak.
*/
abstract predicate isWeak();
}
/**
* A hashing algorithm such as `MD5` or `SHA512`.
*/
class HashingAlgorithm extends MkHashingAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
HashingAlgorithm() { this = MkHashingAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
}
/**
* An encryption algorithm such as `DES` or `AES512`.
*/
class EncryptionAlgorithm extends MkEncryptionAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
EncryptionAlgorithm() { this = MkEncryptionAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
/** Holds if this algorithm is a stream cipher. */
predicate isStreamCipher() { isStreamCipher(name) }
}
/**
* A password hashing algorithm such as `PBKDF2` or `SCRYPT`.
*/
class PasswordHashingAlgorithm extends MkPasswordHashingAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
PasswordHashingAlgorithm() { this = MkPasswordHashingAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
}

2
rust/ql/lib/codeql/rust/security/SensitiveData.qll
View File

@@ -6,7 +6,7 @@
*/ */
import rust import rust
import internal.SensitiveDataHeuristics import codeql.concepts.internal.SensitiveDataHeuristics
private import codeql.rust.dataflow.DataFlow private import codeql.rust.dataflow.DataFlow
/** /**

84
rust/ql/lib/codeql/rust/security/internal/CryptoAlgorithmNames.qll
View File

@@ -1,84 +0,0 @@
/**
* Names of cryptographic algorithms, separated into strong and weak variants.
*
* The names are normalized: upper-case, no spaces, dashes or underscores.
*
* The names are inspired by the names used in real world crypto libraries.
*
* The classification into strong and weak are based on Wikipedia, OWASP and Google (2021).
*/
/**
* Holds if `name` corresponds to a strong hashing algorithm.
*/
predicate isStrongHashingAlgorithm(string name) {
name =
[
// see https://cryptography.io/en/latest/hazmat/primitives/cryptographic-hashes/#blake2
// and https://www.blake2.net/
"BLAKE2", "BLAKE2B", "BLAKE2S",
// see https://github.com/BLAKE3-team/BLAKE3
"BLAKE3",
//
"DSA", "ED25519", "ES256", "ECDSA256", "ES384", "ECDSA384", "ES512", "ECDSA512", "SHA2",
"SHA224", "SHA256", "SHA384", "SHA512", "SHA3", "SHA3224", "SHA3256", "SHA3384", "SHA3512",
// see https://cryptography.io/en/latest/hazmat/primitives/cryptographic-hashes/#cryptography.hazmat.primitives.hashes.SHAKE128
"SHAKE128", "SHAKE256",
// see https://cryptography.io/en/latest/hazmat/primitives/cryptographic-hashes/#sm3
"SM3",
// see https://security.stackexchange.com/a/216297
"WHIRLPOOL",
]
}
/**
* Holds if `name` corresponds to a weak hashing algorithm.
*/
predicate isWeakHashingAlgorithm(string name) {
name =
[
"HAVEL128", "MD2", "MD4", "MD5", "PANAMA", "RIPEMD", "RIPEMD128", "RIPEMD256", "RIPEMD160",
"RIPEMD320", "SHA0", "SHA1"
]
}
/**
* Holds if `name` corresponds to a strong encryption algorithm.
*/
predicate isStrongEncryptionAlgorithm(string name) {
name =
[
"AES", "AES128", "AES192", "AES256", "AES512", "AES-128", "AES-192", "AES-256", "AES-512",
"ARIA", "BLOWFISH", "BF", "ECIES", "CAST", "CAST5", "CAMELLIA", "CAMELLIA128", "CAMELLIA192",
"CAMELLIA256", "CAMELLIA-128", "CAMELLIA-192", "CAMELLIA-256", "CHACHA", "GOST", "GOST89",
"IDEA", "RABBIT", "RSA", "SEED", "SM4"
]
}
/**
* Holds if `name` corresponds to a weak encryption algorithm.
*/
predicate isWeakEncryptionAlgorithm(string name) {
name =
[
"DES", "3DES", "DES3", "TRIPLEDES", "DESX", "TDEA", "TRIPLEDEA", "ARC2", "RC2", "ARC4", "RC4",
"ARCFOUR", "ARC5", "RC5"
]
}
/**
* Holds if `name` corresponds to a strong password hashing algorithm.
*/
predicate isStrongPasswordHashingAlgorithm(string name) {
name = ["ARGON2", "PBKDF2", "BCRYPT", "SCRYPT"]
}
/**
* Holds if `name` corresponds to a weak password hashing algorithm.
*/
predicate isWeakPasswordHashingAlgorithm(string name) { name = "EVPKDF" }
/**
* Holds if `name` corresponds to a stream cipher.
*/
predicate isStreamCipher(string name) { name = ["CHACHA", "RC4", "ARC4", "ARCFOUR", "RABBIT"] }

188
rust/ql/lib/codeql/rust/security/internal/SensitiveDataHeuristics.qll
View File

@@ -1,188 +0,0 @@
/**
* INTERNAL: Do not use.
*
* Provides classes and predicates for identifying strings that may indicate the presence of sensitive data.
* Such that we can share this logic across our CodeQL analysis of different languages.
*
* 'Sensitive' data in general is anything that should not be sent around in unencrypted form.
*/
/**
* A classification of different kinds of sensitive data:
*
* - secret: generic secret or trusted data;
* - id: a user name or other account information;
* - password: a password or authorization key;
* - certificate: a certificate.
* - private: private data such as credit card numbers
*
* While classifications are represented as strings, this should not be relied upon.
* Instead, use the predicates in `SensitiveDataClassification::` to work with
* classifications.
*/
class SensitiveDataClassification extends string {
SensitiveDataClassification() { this in ["secret", "id", "password", "certificate", "private"] }
}
/**
* Provides predicates to select the different kinds of sensitive data we support.
*/
module SensitiveDataClassification {
/** Gets the classification for secret or trusted data. */
SensitiveDataClassification secret() { result = "secret" }
/** Gets the classification for user names or other account information. */
SensitiveDataClassification id() { result = "id" }
/** Gets the classification for passwords or authorization keys. */
SensitiveDataClassification password() { result = "password" }
/** Gets the classification for certificates. */
SensitiveDataClassification certificate() { result = "certificate" }
/** Gets the classification for private data. */
SensitiveDataClassification private() { result = "private" }
}
/**
* INTERNAL: Do not use.
*
* Provides heuristics for identifying names related to sensitive information.
*/
module HeuristicNames {
/**
* Gets a regular expression that identifies strings that may indicate the presence of secret
* or trusted data.
*/
string maybeSecret() { result = "(?is).*((?<!is|is_)secret|(?<!un|un_|is|is_)trusted).*" }
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* user names or other account information.
*/
string maybeAccountInfo() {
result = "(?is).*acc(ou)?nt.*" or
result = "(?is).*(puid|user.?name|user.?id|session.?(id|key)).*" or
result = "(?s).*([uU]|^|_|[a-z](?=U))([uU][iI][dD]).*"
}
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* a password or an authorization key.
*/
string maybePassword() {
result = "(?is).*pass(wd|word|code|.?phrase)(?!.*question).*" or
result = "(?is).*(auth(entication|ori[sz]ation)?).?key.*"
}
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* a certificate.
*/
string maybeCertificate() { result = "(?is).*(cert)(?!.*(format|name|ification)).*" }
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* private data.
*/
string maybePrivate() {
result =
"(?is).*(" +
// Inspired by the list on https://cwe.mitre.org/data/definitions/359.html
// Government identifiers, such as Social Security Numbers
"social.?security|employer.?identification|national.?insurance|resident.?id|" +
"passport.?(num|no)|([_-]|\\b)ssn([_-]|\\b)|" +
// Contact information, such as home addresses
"post.?code|zip.?code|home.?addr|" +
// and telephone numbers
"(mob(ile)?|home).?(num|no|tel|phone)|(tel|fax|phone).?(num|no)|telephone|" +
"emergency.?contact|" +
// Geographic location - where the user is (or was)
"latitude|longitude|nationality|" +
// Financial data - such as credit card numbers, salary, bank accounts, and debts
"(credit|debit|bank|visa).?(card|num|no|acc(ou)?nt)|acc(ou)?nt.?(no|num|credit)|" +
"salary|billing|credit.?(rating|score)|([_-]|\\b)ccn([_-]|\\b)|" +
// Communications - e-mail addresses, private e-mail messages, SMS text messages, chat logs, etc.
// "e(mail|_mail)|" + // this seems too noisy
// Health - medical conditions, insurance status, prescription records
"birth.?da(te|y)|da(te|y).?(of.?)?birth|" +
"medical|(health|care).?plan|healthkit|appointment|prescription|" +
"blood.?(type|alcohol|glucose|pressure)|heart.?(rate|rhythm)|body.?(mass|fat)|" +
"menstrua|pregnan|insulin|inhaler|" +
// Relationships - work and family
"employ(er|ee)|spouse|maiden.?name" +
// ---
").*"
}
/**
* Gets a regular expression that identifies strings that may indicate the presence
* of sensitive data, with `classification` describing the kind of sensitive data involved.
*/
string maybeSensitiveRegexp(SensitiveDataClassification classification) {
result = maybeSecret() and classification = SensitiveDataClassification::secret()
or
result = maybeAccountInfo() and classification = SensitiveDataClassification::id()
or
result = maybePassword() and classification = SensitiveDataClassification::password()
or
result = maybeCertificate() and
classification = SensitiveDataClassification::certificate()
or
result = maybePrivate() and
classification = SensitiveDataClassification::private()
}
/**
* Gets a regular expression that identifies strings that may indicate the presence of data
* that is hashed or encrypted, and hence rendered non-sensitive, or contains special characters
* suggesting nouns within the string do not represent the meaning of the whole string (e.g. a URL or a SQL query).
*
* We also filter out common words like `certain` and `concert`, since otherwise these could
* be matched by the certificate regular expressions. Same for `accountable` (account), or
* `secretarial` (secret).
*/
string notSensitiveRegexp() {
result =
"(?is).*([^\\w$.-]|redact|censor|obfuscate|hash|md5|sha|random|((?<!un)(en))?(crypt|(?<!pass)code)|certain|concert|secretar|accountant|accountab).*"
}
/**
* Holds if `name` may indicate the presence of sensitive data, and `name` does not indicate that
* the data is in fact non-sensitive (for example since it is hashed or encrypted).
*
* That is, one of the regexps from `maybeSensitiveRegexp` matches `name` (with the given
* classification), and none of the regexps from `notSensitiveRegexp` matches `name`.
*/
bindingset[name]
predicate nameIndicatesSensitiveData(string name) {
exists(string combinedRegexp |
// Combine all the maybe-sensitive regexps into one using non-capturing groups and |.
combinedRegexp =
"(?:" + strictconcat(string r | r = maybeSensitiveRegexp(_) | r, ")|(?:") + ")"
|
name.regexpMatch(combinedRegexp)
) and
not name.regexpMatch(notSensitiveRegexp())
}
/**
* Holds if `name` may indicate the presence of sensitive data, and
* `name` does not indicate that the data is in fact non-sensitive (for example since
* it is hashed or encrypted). `classification` describes the kind of sensitive data
* involved.
*
* That is, one of the regexps from `maybeSensitiveRegexp` matches `name` (with the
* given classification), and none of the regexps from `notSensitiveRegexp` matches
* `name`.
*
* When the set of names is large, it's worth using `nameIndicatesSensitiveData/1` as a first
* pass, since that combines all the regexps into one, and should be faster. Then call this
* predicate to get the classification(s).
*/
bindingset[name]
predicate nameIndicatesSensitiveData(string name, SensitiveDataClassification classification) {
name.regexpMatch(maybeSensitiveRegexp(classification)) and
not name.regexpMatch(notSensitiveRegexp())
}
}

1
rust/ql/lib/qlpack.yml
View File

@@ -6,6 +6,7 @@ dbscheme: rust.dbscheme
library: true library: true
upgrades: upgrades upgrades: upgrades
dependencies: dependencies:
codeql/concepts: ${workspace}
codeql/controlflow: ${workspace} codeql/controlflow: ${workspace}
codeql/dataflow: ${workspace} codeql/dataflow: ${workspace}
codeql/regex: ${workspace} codeql/regex: ${workspace}

4
shared/concepts/change-notes/2025-07-16-initial-shared-concepts.md Normal file
View File

@@ -0,0 +1,4 @@
---
category: minorAnalysis
---
* Initial release. Moves the shared concepts library into its own qlpack.

192
shared/concepts/codeql/concepts/ConceptsShared.qll Normal file
View File

@@ -0,0 +1,192 @@
/**
* Provides Concepts which are shared across languages.
*
* Each language has a language specific `Concepts.qll` file that can import the
* shared concepts from this file. A language can either re-export the concept directly,
* or can add additional member-predicates that are needed for that language.
*
* Moving forward, `Concepts.qll` will be the staging ground for brand new concepts from
* each language, but we will maintain a discipline of moving those concepts to
* `ConceptsShared.qll` ASAP.
*/
overlay[local?]
module;
private import CryptoAlgorithms as CA
private import codeql.dataflow.DataFlow as DF
private import codeql.util.Location
/**
* Construct the shared concepts modules.
*/
module ConceptsMake<LocationSig Location, DF::InputSig<Location> DataFlowLang> {
final private class DataFlowNode = DataFlowLang::Node;
/**
* Provides models for cryptographic concepts.
*
* Note: The `CryptographicAlgorithm` class currently doesn't take weak keys into
* consideration for the `isWeak` member predicate. So RSA is always considered
* secure, although using a low number of bits will actually make it insecure. We plan
* to improve our libraries in the future to more precisely capture this aspect.
*/
module Cryptography {
class CryptographicAlgorithm = CA::CryptographicAlgorithm;
class EncryptionAlgorithm = CA::EncryptionAlgorithm;
class HashingAlgorithm = CA::HashingAlgorithm;
class PasswordHashingAlgorithm = CA::PasswordHashingAlgorithm;
/**
* A data flow node that is an application of a cryptographic algorithm. For example,
* encryption, decryption, signature-validation.
*
* Extend this class to refine existing API models. If you want to model new APIs,
* extend `CryptographicOperation::Range` instead.
*/
class CryptographicOperation extends DataFlowNode instanceof CryptographicOperation::Range {
/** Gets the algorithm used, if it matches a known `CryptographicAlgorithm`. */
CryptographicAlgorithm getAlgorithm() { result = super.getAlgorithm() }
/** Gets the data flow node where the cryptographic algorithm used in this operation is configured. */
DataFlowNode getInitialization() { result = super.getInitialization() }
/** Gets an input the algorithm is used on, for example the plain text input to be encrypted. */
DataFlowNode getAnInput() { result = super.getAnInput() }
/**
* Gets the block mode used to perform this cryptographic operation.
*
* This predicate is only expected to have a result if two conditions hold:
* 1. The operation is an encryption operation, i.e. the algorithm used is an `EncryptionAlgorithm`, and
* 2. The algorithm used is a block cipher (not a stream cipher).
*
* If either of these conditions do not hold, then this predicate should have no result.
*/
BlockMode getBlockMode() { result = super.getBlockMode() }
}
/** Provides classes for modeling new applications of a cryptographic algorithms. */
module CryptographicOperation {
/**
* A data flow node that is an application of a cryptographic algorithm. For example,
* encryption, decryption, signature-validation.
*
* Extend this class to model new APIs. If you want to refine existing API models,
* extend `CryptographicOperation` instead.
*/
abstract class Range extends DataFlowNode {
/** Gets the data flow node where the cryptographic algorithm used in this operation is configured. */
abstract DataFlowNode getInitialization();
/** Gets the algorithm used, if it matches a known `CryptographicAlgorithm`. */
abstract CryptographicAlgorithm getAlgorithm();
/** Gets an input the algorithm is used on, for example the plain text input to be encrypted. */
abstract DataFlowNode getAnInput();
/**
* Gets the block mode used to perform this cryptographic operation.
*
* This predicate is only expected to have a result if two conditions hold:
* 1. The operation is an encryption operation, i.e. the algorithm used is an `EncryptionAlgorithm`, and
* 2. The algorithm used is a block cipher (not a stream cipher).
*
* If either of these conditions do not hold, then this predicate should have no result.
*/
abstract BlockMode getBlockMode();
}
}
/**
* A cryptographic block cipher mode of operation. This can be used to encrypt
* data of arbitrary length using a block encryption algorithm.
*/
class BlockMode extends string {
BlockMode() {
this =
[
"ECB", "CBC", "GCM", "CCM", "CFB", "OFB", "CTR", "OPENPGP",
"XTS", // https://csrc.nist.gov/publications/detail/sp/800-38e/final
"EAX" // https://en.wikipedia.org/wiki/EAX_mode
]
}
/** Holds if this block mode is considered to be insecure. */
predicate isWeak() { this = "ECB" }
/** Holds if the given string appears to match this block mode. */
bindingset[s]
predicate matchesString(string s) { s.toUpperCase().matches("%" + this + "%") }
}
}
/** Provides classes for modeling HTTP-related APIs. */
module Http {
/** Provides classes for modeling HTTP clients. */
module Client {
/**
* A data flow node that makes an outgoing HTTP request.
*
* Extend this class to refine existing API models. If you want to model new APIs,
* extend `Http::Client::Request::Range` instead.
*/
class Request extends DataFlowNode instanceof Request::Range {
/**
* Gets a data flow node that contributes to the URL of the request.
* Depending on the framework, a request may have multiple nodes which contribute to the URL.
*/
DataFlowNode getAUrlPart() { result = super.getAUrlPart() }
/** Gets a string that identifies the framework used for this request. */
string getFramework() { result = super.getFramework() }
/**
* Holds if this request is made using a mode that disables SSL/TLS
* certificate validation, where `disablingNode` represents the point at
* which the validation was disabled, and `argumentOrigin` represents the origin
* of the argument that disabled the validation (which could be the same node as
* `disablingNode`).
*/
predicate disablesCertificateValidation(
DataFlowNode disablingNode, DataFlowNode argumentOrigin
) {
super.disablesCertificateValidation(disablingNode, argumentOrigin)
}
}
/** Provides a class for modeling new HTTP requests. */
module Request {
/**
* A data flow node that makes an outgoing HTTP request.
*
* Extend this class to model new APIs. If you want to refine existing API models,
* extend `Http::Client::Request` instead.
*/
abstract class Range extends DataFlowNode {
/**
* Gets a data flow node that contributes to the URL of the request.
* Depending on the framework, a request may have multiple nodes which contribute to the URL.
*/
abstract DataFlowNode getAUrlPart();
/** Gets a string that identifies the framework used for this request. */
abstract string getFramework();
/**
* Holds if this request is made using a mode that disables SSL/TLS
* certificate validation, where `disablingNode` represents the point at
* which the validation was disabled, and `argumentOrigin` represents the origin
* of the argument that disabled the validation (which could be the same node as
* `disablingNode`).
*/
abstract predicate disablesCertificateValidation(
DataFlowNode disablingNode, DataFlowNode argumentOrigin
);
}
}
}
}
}

119
shared/concepts/codeql/concepts/CryptoAlgorithms.qll Normal file
View File

@@ -0,0 +1,119 @@
/**
* Provides classes modeling cryptographic algorithms, separated into strong and weak variants.
*
* The classification into strong and weak are based on Wikipedia, OWASP and Google (2021).
*/
overlay[local?]
module;
private import codeql.concepts.internal.CryptoAlgorithmNames
/**
* A cryptographic algorithm.
*/
private newtype TCryptographicAlgorithm =
MkHashingAlgorithm(string name, boolean isWeak) {
isStrongHashingAlgorithm(name) and isWeak = false
or
isWeakHashingAlgorithm(name) and isWeak = true
} or
MkEncryptionAlgorithm(string name, boolean isWeak) {
isStrongEncryptionAlgorithm(name) and isWeak = false
or
isWeakEncryptionAlgorithm(name) and isWeak = true
} or
MkPasswordHashingAlgorithm(string name, boolean isWeak) {
isStrongPasswordHashingAlgorithm(name) and isWeak = false
or
isWeakPasswordHashingAlgorithm(name) and isWeak = true
}
/**
* Gets the most specific `CryptographicAlgorithm` that matches the given `name`.
* A matching algorithm is one where the name of the algorithm matches the start of name, with allowances made for different name formats.
* In the case that multiple `CryptographicAlgorithm`s match the given `name`, the algorithm(s) with the longest name will be selected. This is intended to select more specific versions of algorithms when multiple versions could match - for example "SHA3_224" matches against both "SHA3" and "SHA3224", but the latter is a more precise match.
*/
bindingset[name]
private CryptographicAlgorithm getBestAlgorithmForName(string name) {
result =
max(CryptographicAlgorithm algorithm |
algorithm.getName() =
[
name.toUpperCase(), // the full name
name.toUpperCase().regexpCapture("^([\\w]+)(?:-.*)?$", 1), // the name prior to any dashes or spaces
name.toUpperCase().regexpCapture("^([A-Z0-9]+)(?:(-|_).*)?$", 1) // the name prior to any dashes, spaces, or underscores
].regexpReplaceAll("[-_ ]", "") // strip dashes, underscores, and spaces
|
algorithm order by algorithm.getName().length()
)
}
/**
* A cryptographic algorithm.
*/
abstract class CryptographicAlgorithm extends TCryptographicAlgorithm {
/** Gets a textual representation of this element. */
string toString() { result = this.getName() }
/**
* Gets the normalized name of this algorithm (upper-case, no spaces, dashes or underscores).
*/
abstract string getName();
/**
* Holds if the name of this algorithm is the most specific match for `name`.
* This predicate matches quite liberally to account for different ways of formatting algorithm names, e.g. using dashes, underscores, or spaces as separators, including or not including block modes of operation, etc.
*/
bindingset[name]
predicate matchesName(string name) { this = getBestAlgorithmForName(name) }
/**
* Holds if this algorithm is weak.
*/
abstract predicate isWeak();
}
/**
* A hashing algorithm such as `MD5` or `SHA512`.
*/
class HashingAlgorithm extends MkHashingAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
HashingAlgorithm() { this = MkHashingAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
}
/**
* An encryption algorithm such as `DES` or `AES512`.
*/
class EncryptionAlgorithm extends MkEncryptionAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
EncryptionAlgorithm() { this = MkEncryptionAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
/** Holds if this algorithm is a stream cipher. */
predicate isStreamCipher() { isStreamCipher(name) }
}
/**
* A password hashing algorithm such as `PBKDF2` or `SCRYPT`.
*/
class PasswordHashingAlgorithm extends MkPasswordHashingAlgorithm, CryptographicAlgorithm {
string name;
boolean isWeak;
PasswordHashingAlgorithm() { this = MkPasswordHashingAlgorithm(name, isWeak) }
override string getName() { result = name }
override predicate isWeak() { isWeak = true }
}

2
python/ql/lib/semmle/python/concepts/internal/CryptoAlgorithmNames.qll → shared/concepts/codeql/concepts/internal/CryptoAlgorithmNames.qll
View File

@@ -7,6 +7,8 @@
* *
* The classification into strong and weak are based on Wikipedia, OWASP and Google (2021). * The classification into strong and weak are based on Wikipedia, OWASP and Google (2021).
*/ */
overlay[local?]
module;
/** /**
* Holds if `name` corresponds to a strong hashing algorithm. * Holds if `name` corresponds to a strong hashing algorithm.

2
ruby/ql/lib/codeql/ruby/security/internal/SensitiveDataHeuristics.qll → shared/concepts/codeql/concepts/internal/SensitiveDataHeuristics.qll
View File

@@ -6,6 +6,8 @@
* *
* 'Sensitive' data in general is anything that should not be sent around in unencrypted form. * 'Sensitive' data in general is anything that should not be sent around in unencrypted form.
*/ */
overlay[local?]
module;
/** /**
* A classification of different kinds of sensitive data: * A classification of different kinds of sensitive data:

8
shared/concepts/qlpack.yml Normal file
View File

@@ -0,0 +1,8 @@
name: codeql/concepts
version: 0.0.0-dev
groups: shared
library: true
dependencies:
codeql/dataflow: ${workspace}
codeql/util: ${workspace}
warnOnImplicitThis: true

2
swift/ql/lib/codeql/swift/security/SensitiveExprs.qll
View File

@@ -5,7 +5,7 @@
*/ */
import swift import swift
import internal.SensitiveDataHeuristics import codeql.concepts.internal.SensitiveDataHeuristics
private import codeql.swift.dataflow.DataFlow private import codeql.swift.dataflow.DataFlow
private import codeql.swift.dataflow.ExternalFlow private import codeql.swift.dataflow.ExternalFlow

188
swift/ql/lib/codeql/swift/security/internal/SensitiveDataHeuristics.qll
View File

@@ -1,188 +0,0 @@
/**
* INTERNAL: Do not use.
*
* Provides classes and predicates for identifying strings that may indicate the presence of sensitive data.
* Such that we can share this logic across our CodeQL analysis of different languages.
*
* 'Sensitive' data in general is anything that should not be sent around in unencrypted form.
*/
/**
* A classification of different kinds of sensitive data:
*
* - secret: generic secret or trusted data;
* - id: a user name or other account information;
* - password: a password or authorization key;
* - certificate: a certificate.
* - private: private data such as credit card numbers
*
* While classifications are represented as strings, this should not be relied upon.
* Instead, use the predicates in `SensitiveDataClassification::` to work with
* classifications.
*/
class SensitiveDataClassification extends string {
SensitiveDataClassification() { this in ["secret", "id", "password", "certificate", "private"] }
}
/**
* Provides predicates to select the different kinds of sensitive data we support.
*/
module SensitiveDataClassification {
/** Gets the classification for secret or trusted data. */
SensitiveDataClassification secret() { result = "secret" }
/** Gets the classification for user names or other account information. */
SensitiveDataClassification id() { result = "id" }
/** Gets the classification for passwords or authorization keys. */
SensitiveDataClassification password() { result = "password" }
/** Gets the classification for certificates. */
SensitiveDataClassification certificate() { result = "certificate" }
/** Gets the classification for private data. */
SensitiveDataClassification private() { result = "private" }
}
/**
* INTERNAL: Do not use.
*
* Provides heuristics for identifying names related to sensitive information.
*/
module HeuristicNames {
/**
* Gets a regular expression that identifies strings that may indicate the presence of secret
* or trusted data.
*/
string maybeSecret() { result = "(?is).*((?<!is|is_)secret|(?<!un|un_|is|is_)trusted).*" }
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* user names or other account information.
*/
string maybeAccountInfo() {
result = "(?is).*acc(ou)?nt.*" or
result = "(?is).*(puid|user.?name|user.?id|session.?(id|key)).*" or
result = "(?s).*([uU]|^|_|[a-z](?=U))([uU][iI][dD]).*"
}
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* a password or an authorization key.
*/
string maybePassword() {
result = "(?is).*pass(wd|word|code|.?phrase)(?!.*question).*" or
result = "(?is).*(auth(entication|ori[sz]ation)?).?key.*"
}
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* a certificate.
*/
string maybeCertificate() { result = "(?is).*(cert)(?!.*(format|name|ification)).*" }
/**
* Gets a regular expression that identifies strings that may indicate the presence of
* private data.
*/
string maybePrivate() {
result =
"(?is).*(" +
// Inspired by the list on https://cwe.mitre.org/data/definitions/359.html
// Government identifiers, such as Social Security Numbers
"social.?security|employer.?identification|national.?insurance|resident.?id|" +
"passport.?(num|no)|([_-]|\\b)ssn([_-]|\\b)|" +
// Contact information, such as home addresses
"post.?code|zip.?code|home.?addr|" +
// and telephone numbers
"(mob(ile)?|home).?(num|no|tel|phone)|(tel|fax|phone).?(num|no)|telephone|" +
"emergency.?contact|" +
// Geographic location - where the user is (or was)
"latitude|longitude|nationality|" +
// Financial data - such as credit card numbers, salary, bank accounts, and debts
"(credit|debit|bank|visa).?(card|num|no|acc(ou)?nt)|acc(ou)?nt.?(no|num|credit)|" +
"salary|billing|credit.?(rating|score)|([_-]|\\b)ccn([_-]|\\b)|" +
// Communications - e-mail addresses, private e-mail messages, SMS text messages, chat logs, etc.
// "e(mail|_mail)|" + // this seems too noisy
// Health - medical conditions, insurance status, prescription records
"birth.?da(te|y)|da(te|y).?(of.?)?birth|" +
"medical|(health|care).?plan|healthkit|appointment|prescription|" +
"blood.?(type|alcohol|glucose|pressure)|heart.?(rate|rhythm)|body.?(mass|fat)|" +
"menstrua|pregnan|insulin|inhaler|" +
// Relationships - work and family
"employ(er|ee)|spouse|maiden.?name" +
// ---
").*"
}
/**
* Gets a regular expression that identifies strings that may indicate the presence
* of sensitive data, with `classification` describing the kind of sensitive data involved.
*/
string maybeSensitiveRegexp(SensitiveDataClassification classification) {
result = maybeSecret() and classification = SensitiveDataClassification::secret()
or
result = maybeAccountInfo() and classification = SensitiveDataClassification::id()
or
result = maybePassword() and classification = SensitiveDataClassification::password()
or
result = maybeCertificate() and
classification = SensitiveDataClassification::certificate()
or
result = maybePrivate() and
classification = SensitiveDataClassification::private()
}
/**
* Gets a regular expression that identifies strings that may indicate the presence of data
* that is hashed or encrypted, and hence rendered non-sensitive, or contains special characters
* suggesting nouns within the string do not represent the meaning of the whole string (e.g. a URL or a SQL query).
*
* We also filter out common words like `certain` and `concert`, since otherwise these could
* be matched by the certificate regular expressions. Same for `accountable` (account), or
* `secretarial` (secret).
*/
string notSensitiveRegexp() {
result =
"(?is).*([^\\w$.-]|redact|censor|obfuscate|hash|md5|sha|random|((?<!un)(en))?(crypt|(?<!pass)code)|certain|concert|secretar|accountant|accountab).*"
}
/**
* Holds if `name` may indicate the presence of sensitive data, and `name` does not indicate that
* the data is in fact non-sensitive (for example since it is hashed or encrypted).
*
* That is, one of the regexps from `maybeSensitiveRegexp` matches `name` (with the given
* classification), and none of the regexps from `notSensitiveRegexp` matches `name`.
*/
bindingset[name]
predicate nameIndicatesSensitiveData(string name) {
exists(string combinedRegexp |
// Combine all the maybe-sensitive regexps into one using non-capturing groups and |.
combinedRegexp =
"(?:" + strictconcat(string r | r = maybeSensitiveRegexp(_) | r, ")|(?:") + ")"
|
name.regexpMatch(combinedRegexp)
) and
not name.regexpMatch(notSensitiveRegexp())
}
/**
* Holds if `name` may indicate the presence of sensitive data, and
* `name` does not indicate that the data is in fact non-sensitive (for example since
* it is hashed or encrypted). `classification` describes the kind of sensitive data
* involved.
*
* That is, one of the regexps from `maybeSensitiveRegexp` matches `name` (with the
* given classification), and none of the regexps from `notSensitiveRegexp` matches
* `name`.
*
* When the set of names is large, it's worth using `nameIndicatesSensitiveData/1` as a first
* pass, since that combines all the regexps into one, and should be faster. Then call this
* predicate to get the classification(s).
*/
bindingset[name]
predicate nameIndicatesSensitiveData(string name, SensitiveDataClassification classification) {
name.regexpMatch(maybeSensitiveRegexp(classification)) and
not name.regexpMatch(notSensitiveRegexp())
}
}

1
swift/ql/lib/qlpack.yml
View File

@@ -6,6 +6,7 @@ dbscheme: swift.dbscheme
upgrades: upgrades upgrades: upgrades
library: true library: true
dependencies: dependencies:
codeql/concepts: ${workspace}
codeql/controlflow: ${workspace} codeql/controlflow: ${workspace}
codeql/dataflow: ${workspace} codeql/dataflow: ${workspace}
codeql/regex: ${workspace} codeql/regex: ${workspace}