Go: Cleanup and renaming.

This commit is contained in:
Michael Nebel
2022-12-19 15:43:49 +01:00
parent 3749a1bd4d
commit 48d0eccbf6
9 changed files with 65 additions and 133 deletions

View File

@@ -1,19 +1,20 @@
/**
* INTERNAL use only. This is an experimental API subject to change without notice.
*
* Provides classes and predicates for dealing with flow models specified in CSV format.
* Provides classes and predicates for dealing with MaD flow models specified
* in data extensions and CSV format.
*
* The CSV specification has the following columns:
* - Sources:
* `namespace; type; subtypes; name; signature; ext; output; kind`
* `package; type; subtypes; name; signature; ext; output; kind; provenance`
* - Sinks:
* `namespace; type; subtypes; name; signature; ext; input; kind`
* `package; type; subtypes; name; signature; ext; input; kind; provenance`
* - Summaries:
* `namespace; type; subtypes; name; signature; ext; input; output; kind`
* `package; type; subtypes; name; signature; ext; input; output; kind; provenance`
*
* The interpretation of a row is similar to API-graphs with a left-to-right
* reading.
* 1. The `namespace` column selects a package.
* 1. The `package` column selects a package.
* 2. The `type` column selects a type within that package.
* 3. The `subtypes` is a boolean that indicates whether to jump to an
* arbitrary subtype of that type.
@@ -76,45 +77,6 @@ private module Frameworks {
private import semmle.go.frameworks.Stdlib
}
/**
* A unit class for adding additional source model rows.
*
* Extend this class to add additional source definitions.
*/
class SourceModelCsv extends Unit {
/** Holds if `row` specifies a source definition. */
abstract predicate row(string row);
}
/**
* A unit class for adding additional sink model rows.
*
* Extend this class to add additional sink definitions.
*/
class SinkModelCsv extends Unit {
/** Holds if `row` specifies a sink definition. */
abstract predicate row(string row);
}
/**
* A unit class for adding additional summary model rows.
*
* Extend this class to add additional flow summary definitions.
*/
class SummaryModelCsv extends Unit {
/** Holds if `row` specifies a summary definition. */
abstract predicate row(string row);
}
/** Holds if `row` is a source model. */
predicate sourceModel(string row) { any(SourceModelCsv s).row(row) }
/** Holds if `row` is a sink model. */
predicate sinkModel(string row) { any(SinkModelCsv s).row(row) }
/** Holds if `row` is a summary model. */
predicate summaryModel(string row) { any(SummaryModelCsv s).row(row) }
/** Holds if a source model exists for the given parameters. */
predicate sourceModel = Extensions::sourceModel/9;
@@ -124,35 +86,35 @@ predicate sinkModel = Extensions::sinkModel/9;
/** Holds if a summary model exists for the given parameters. */
predicate summaryModel = Extensions::summaryModel/10;
/** Holds if `package` have CSV framework coverage. */
private predicate packageHasCsvCoverage(string package) {
/** Holds if `package` have MaD framework coverage. */
private predicate packageHasMaDCoverage(string package) {
sourceModel(package, _, _, _, _, _, _, _, _) or
sinkModel(package, _, _, _, _, _, _, _, _) or
summaryModel(package, _, _, _, _, _, _, _, _, _)
}
/**
* Holds if `package` and `subpkg` have CSV framework coverage and `subpkg`
* Holds if `package` and `subpkg` have MaD framework coverage and `subpkg`
* is a subpackage of `package`.
*/
private predicate packageHasASubpackage(string package, string subpkg) {
packageHasCsvCoverage(package) and
packageHasCsvCoverage(subpkg) and
packageHasMaDCoverage(package) and
packageHasMaDCoverage(subpkg) and
subpkg.prefix(subpkg.indexOf(".")) = package
}
/**
* Holds if `package` has CSV framework coverage and it is not a subpackage of
* any other package with CSV framework coverage.
* Holds if `package` has MaD framework coverage and it is not a subpackage of
* any other package with MaD framework coverage.
*/
private predicate canonicalPackage(string package) {
packageHasCsvCoverage(package) and not packageHasASubpackage(_, package)
packageHasMaDCoverage(package) and not packageHasASubpackage(_, package)
}
/**
* Holds if `package` and `subpkg` have CSV framework coverage, `subpkg` is a
* Holds if `package` and `subpkg` have MaD framework coverage, `subpkg` is a
* subpackage of `package` (or they are the same), and `package` is not a
* subpackage of any other package with CSV framework coverage.
* subpackage of any other package with MaD framework coverage.
*/
private predicate canonicalPackageHasASubpackage(string package, string subpkg) {
canonicalPackage(package) and
@@ -160,9 +122,9 @@ private predicate canonicalPackageHasASubpackage(string package, string subpkg)
}
/**
* Holds if CSV framework coverage of `package` is `n` api endpoints of the
* Holds if MaD framework coverage of `package` is `n` api endpoints of the
* kind `(kind, part)`, and `pkgs` is the number of subpackages of `package`
* which have CSV framework coverage (including `package` itself).
* which have MaD framework coverage (including `package` itself).
*/
predicate modelCoverage(string package, int pkgs, string kind, string part, int n) {
pkgs = strictcount(string subpkg | canonicalPackageHasASubpackage(package, subpkg)) and
@@ -193,8 +155,8 @@ predicate modelCoverage(string package, int pkgs, string kind, string part, int
)
}
/** Provides a query predicate to check the CSV data for validation errors. */
module CsvValidation {
/** Provides a query predicate to check the MaD models for validation errors. */
module ModelValidation {
private string getInvalidModelInput() {
exists(string pred, AccessPath input, string part |
sinkModel(_, _, _, _, _, _, input, _, _) and pred = "sink"
@@ -227,57 +189,25 @@ module CsvValidation {
}
private string getInvalidModelKind() {
exists(string row, string kind | summaryModel(row) |
kind = row.splitAt(";", 8) and
exists(string kind | summaryModel(_, _, _, _, _, _, _, _, kind, _) |
not kind = ["taint", "value"] and
result = "Invalid kind \"" + kind + "\" in summary model."
)
}
private string getInvalidModelSubtype() {
exists(string pred, string row |
sourceModel(row) and pred = "source"
or
sinkModel(row) and pred = "sink"
or
summaryModel(row) and pred = "summary"
|
exists(string b |
b = row.splitAt(";", 2) and
not b = ["true", "false"] and
result = "Invalid boolean \"" + b + "\" in " + pred + " model."
)
)
}
private string getInvalidModelColumnCount() {
exists(string pred, string row, int expect |
sourceModel(row) and expect = 8 and pred = "source"
or
sinkModel(row) and expect = 8 and pred = "sink"
or
summaryModel(row) and expect = 9 and pred = "summary"
|
exists(int cols |
cols = 1 + max(int n | exists(row.splitAt(";", n))) and
cols != expect and
result =
"Wrong number of columns in " + pred + " model row, expected " + expect + ", got " + cols +
"."
)
)
}
private string getInvalidModelSignature() {
exists(string pred, string namespace, string type, string name, string signature, string ext |
sourceModel(namespace, type, _, name, signature, ext, _, _, _) and pred = "source"
or
sinkModel(namespace, type, _, name, signature, ext, _, _, _) and pred = "sink"
or
summaryModel(namespace, type, _, name, signature, ext, _, _, _, _) and pred = "summary"
exists(
string pred, string package, string type, string name, string signature, string ext,
string provenance
|
not namespace.regexpMatch("[a-zA-Z0-9_\\./]*") and
result = "Dubious namespace \"" + namespace + "\" in " + pred + " model."
sourceModel(package, type, _, name, signature, ext, _, _, provenance) and pred = "source"
or
sinkModel(package, type, _, name, signature, ext, _, _, provenance) and pred = "sink"
or
summaryModel(package, type, _, name, signature, ext, _, _, _, provenance) and pred = "summary"
|
not package.regexpMatch("[a-zA-Z0-9_\\./]*") and
result = "Dubious package \"" + package + "\" in " + pred + " model."
or
not type.regexpMatch("[a-zA-Z0-9_\\$<>]*") and
result = "Dubious type \"" + type + "\" in " + pred + " model."
@@ -290,26 +220,29 @@ module CsvValidation {
or
not ext.regexpMatch("|Annotated") and
result = "Unrecognized extra API graph element \"" + ext + "\" in " + pred + " model."
or
not provenance = ["manual", "generated"] and
result = "Unrecognized provenance description \"" + provenance + "\" in " + pred + " model."
)
}
/** Holds if some row in a CSV-based flow model appears to contain typos. */
/** Holds if some row in a MaD flow model appears to contain typos. */
query predicate invalidModelRow(string msg) {
msg =
[
getInvalidModelSignature(), getInvalidModelInput(), getInvalidModelOutput(),
getInvalidModelSubtype(), getInvalidModelColumnCount(), getInvalidModelKind()
getInvalidModelKind()
]
}
}
pragma[nomagic]
private predicate elementSpec(
string namespace, string type, boolean subtypes, string name, string signature, string ext
string package, string type, boolean subtypes, string name, string signature, string ext
) {
sourceModel(namespace, type, subtypes, name, signature, ext, _, _, _) or
sinkModel(namespace, type, subtypes, name, signature, ext, _, _, _) or
summaryModel(namespace, type, subtypes, name, signature, ext, _, _, _, _)
sourceModel(package, type, subtypes, name, signature, ext, _, _, _) or
sinkModel(package, type, subtypes, name, signature, ext, _, _, _) or
summaryModel(package, type, subtypes, name, signature, ext, _, _, _, _)
}
private string paramsStringPart(Function f, int i) {
@@ -405,7 +338,7 @@ predicate parseContent(string component, DataFlow::Content content) {
cached
private module Cached {
/**
* Holds if `node` is specified as a source with the given kind in a CSV flow
* Holds if `node` is specified as a source with the given kind in a MaD flow
* model.
*/
cached
@@ -414,7 +347,7 @@ private module Cached {
}
/**
* Holds if `node` is specified as a sink with the given kind in a CSV flow
* Holds if `node` is specified as a sink with the given kind in a MaD flow
* model.
*/
cached

View File

@@ -65,11 +65,9 @@ DataFlowType getSyntheticGlobalType(SummaryComponent::SyntheticGlobal sg) { any(
predicate summaryElement(
SummarizedCallableBase c, string input, string output, string kind, string provenance
) {
exists(
string namespace, string type, boolean subtypes, string name, string signature, string ext
|
summaryModel(namespace, type, subtypes, name, signature, ext, input, output, kind, provenance) and
c.asFunction() = interpretElement(namespace, type, subtypes, name, signature, ext).asEntity()
exists(string package, string type, boolean subtypes, string name, string signature, string ext |
summaryModel(package, type, subtypes, name, signature, ext, input, output, kind, provenance) and
c.asFunction() = interpretElement(package, type, subtypes, name, signature, ext).asEntity()
)
}
@@ -154,11 +152,9 @@ class SourceOrSinkElement extends TSourceOrSinkElement {
* `output`, kind `kind`, and provenance `provenance`.
*/
predicate sourceElement(SourceOrSinkElement e, string output, string kind, string provenance) {
exists(
string namespace, string type, boolean subtypes, string name, string signature, string ext
|
sourceModel(namespace, type, subtypes, name, signature, ext, output, kind, provenance) and
e = interpretElement(namespace, type, subtypes, name, signature, ext)
exists(string package, string type, boolean subtypes, string name, string signature, string ext |
sourceModel(package, type, subtypes, name, signature, ext, output, kind, provenance) and
e = interpretElement(package, type, subtypes, name, signature, ext)
)
}
@@ -167,11 +163,9 @@ predicate sourceElement(SourceOrSinkElement e, string output, string kind, strin
* `input`, kind `kind` and provenance `provenance`.
*/
predicate sinkElement(SourceOrSinkElement e, string input, string kind, string provenance) {
exists(
string namespace, string type, boolean subtypes, string name, string signature, string ext
|
sinkModel(namespace, type, subtypes, name, signature, ext, input, kind, provenance) and
e = interpretElement(namespace, type, subtypes, name, signature, ext)
exists(string package, string type, boolean subtypes, string name, string signature, string ext |
sinkModel(package, type, subtypes, name, signature, ext, input, kind, provenance) and
e = interpretElement(package, type, subtypes, name, signature, ext)
)
}

View File

@@ -23,7 +23,12 @@ module UntrustedFlowSource {
*/
abstract class Range extends DataFlow::Node { }
class CsvRemoteSource extends Range {
CsvRemoteSource() { ExternalFlow::sourceNode(this, "remote") }
/**
* A source of data that is controlled by an untrusted user.
*/
class MaDRemoteSource extends Range {
MaDRemoteSource() { ExternalFlow::sourceNode(this, "remote") }
}
deprecated class CsvRemoteSource = MaDRemoteSource;
}

View File

@@ -4,7 +4,7 @@
import go
import semmle.go.dataflow.ExternalFlow
import CsvValidation
import ModelValidation
import semmle.go.dataflow.internal.FlowSummaryImpl as FlowSummaryImpl
import TestUtilities.InlineFlowTest

View File

@@ -1,6 +1,6 @@
import go
import semmle.go.dataflow.ExternalFlow
import CsvValidation
import ModelValidation
from DataFlow::Node node, string kind
where sinkNode(node, kind)

View File

@@ -1,6 +1,6 @@
import go
import semmle.go.dataflow.ExternalFlow
import CsvValidation
import ModelValidation
from DataFlow::Node node, string kind
where sourceNode(node, kind)

View File

@@ -1,6 +1,6 @@
import go
import semmle.go.dataflow.ExternalFlow
import CsvValidation
import ModelValidation
import semmle.go.dataflow.internal.FlowSummaryImpl as FlowSummaryImpl
from DataFlow::Node node1, DataFlow::Node node2

View File

@@ -1,6 +1,6 @@
import go
import semmle.go.dataflow.ExternalFlow
import CsvValidation
import ModelValidation
import TestUtilities.InlineExpectationsTest
class DataConfiguration extends DataFlow::Configuration {

View File

@@ -1,6 +1,6 @@
import go
import semmle.go.dataflow.ExternalFlow
import CsvValidation
import ModelValidation
// import DataFlow::PartialPathGraph
import TestUtilities.InlineExpectationsTest