Issues

Jinja2 templates (.j2) are invisible to detect — the graph shows generated output but hides its source

#3427

· AmberProof opened

on Sep 8, 2026

detect_incremental derives deleted_files from scan membership, not existence — save_manifest has the exists() check that its own comment says detect_incremental needs, and the placement is backwards in both directions

#3426

· John-kibe opened

on Sep 8, 2026

PHP: use A\B\C as D gives the import edge a target derived from the last segment, splitting the class identity

#3421

· DmitrySamushchenko opened

on Sep 8, 2026

Bash extractor: exec-position paths built from a variable (`"$script_dir/x.sh"`) produce no cross-file edge — the #1756 fix only resolves literal paths

#3416

· edwardselby opened

on Sep 8, 2026

global add and merge-graphs produce different graphs from the same inputs

#3414

· xiongjianxu opened

on Sep 8, 2026

Github diff too large for commited graphify-out content

#3413

· rudifus-ppy opened

on Sep 8, 2026

#3203 unverified-semantic-shrink guard has no verification path: a legitimately edited doc that yields fewer nodes fails every incremental run when the graph is net smaller (re-run reproduces it; only --allow-partial clears it)

#3412

· janmaticka opened

on Sep 8, 2026

Incremental extract wipes PackageReference/TargetFramework nodes and edges of every .csproj referenced from a re-extracted .sln / .csproj (cross-file stub nodes pollute build_merge's replace-set)

#3411

· janmaticka opened

on Sep 8, 2026

PHP: closures produce no nodes in any position — assigned, arrow, or passed as an argument

#3409

· laurism-openelms opened

on Sep 8, 2026

JS: `this.X = function` is extracted only inside a function declaration — every function-expression form drops it (incomplete fix of #1322)

#3408

· laurism-openelms opened

on Sep 8, 2026

Symbol node IDs collide across directories; walk order decides the winner, so output is not reproducible across filesystems

#3407

· fracasamax opened

on Sep 8, 2026

C#: an unqualified generic call (`Get<T>(...)`, `this.Get<T>(...)`) drops its `calls` edge

#3406

· JensD-git opened

on Sep 8, 2026

Navigation Menu

Uh oh!

Uh oh!

All issues

Jinja2 templates (.j2) are invisible to detect — the graph shows generated output but hides its source

detect_incremental derives deleted_files from scan membership, not existence — save_manifest has the exists() check that its own comment says detect_incremental needs, and the placement is backwards in both directions

PHP: use A\B\C as D gives the import edge a target derived from the last segment, splitting the class identity

Bash extractor: exec-position paths built from a variable (`"$script_dir/x.sh"`) produce no cross-file edge — the #1756 fix only resolves literal paths

global add and merge-graphs produce different graphs from the same inputs

Github diff too large for commited graphify-out content

#3203 unverified-semantic-shrink guard has no verification path: a legitimately edited doc that yields fewer nodes fails every incremental run when the graph is net smaller (re-run reproduces it; only --allow-partial clears it)

Incremental extract wipes PackageReference/TargetFramework nodes and edges of every .csproj referenced from a re-extracted .sln / .csproj (cross-file stub nodes pollute build_merge's replace-set)

PHP: closures produce no nodes in any position — assigned, arrow, or passed as an argument

JS: `this.X = function` is extracted only inside a function declaration — every function-expression form drops it (incomplete fix of #1322)

Symbol node IDs collide across directories; walk order decides the winner, so output is not reproducible across filesystems

C#: an unqualified generic call (`Get<T>(...)`, `this.Get<T>(...)`) drops its `calls` edge

Uh oh!

Uh oh!

Search results