chore(deps)(deps): bump the go-dependencies group with 2 updates

[dependabot]

Bumps the go-dependencies group with 2 updates: github.com/bmatcuk/doublestar/v4 and github.com/klippa-app/go-pdfium.

Updates github.com/bmatcuk/doublestar/v4 from 4.10.0 to 4.10.2

Release notes

Sourced from github.com/bmatcuk/doublestar/v4's releases.

Better support for Windows "Junctions"

Has improved support for reading "junctions" in Windows (ie, Windows' version of symlinks).

Thanks @​x0Lazarus for the PR!

What's Changed

• Fix Windows junction traversal with Go 1.23 and later by @​x0Lazarus in bmatcuk/doublestar#117

New Contributors

• @​x0Lazarus made their first contribution in bmatcuk/doublestar#117

Full Changelog: bmatcuk/doublestar@v4.10.1...v4.10.2

Fixed issue with alternatives and character classes

There was an issue where commas or ending curly brackets inside a character class would cause code looking for the next alternative (commas) or the end of alternatives (ending curly brackets) to erroneously match those characters inside the character classes.

Thanks to @​sidsri14 for the fix!

What's Changed

• fix brace alternation scanning with character classes by @​sidsri14 in bmatcuk/doublestar#115

New Contributors

• @​sidsri14 made their first contribution in bmatcuk/doublestar#115

Full Changelog: bmatcuk/doublestar@v4.10.0...v4.10.1

Commits

• fc7b711 update CI
• 7b00623 Merge branch 'x0Lazarus-fix/windows-junction-traversal'
• 146da9c small test changes
• dbb13b8 fix brace alternation scanning with character classes
• 314a609 Merge branch 'sidsri14-master'
• f0902d5 small test changes
• ee530d0 Fix traversal of Windows junctions reported as irregular files
• bf0b1d6 fix brace alternation scanning with character classes
• See full diff in compare view

Updates github.com/klippa-app/go-pdfium from 1.20.0 to 1.20.3

Release notes

Sourced from github.com/klippa-app/go-pdfium's releases.

v1.20.3

What's Changed

• Upgrade pdfium to 8057 by @​jerbob92 in klippa-app/go-pdfium#372

Full Changelog: klippa-app/go-pdfium@v1.20.2...v1.20.3

PDFium 8057

This release updates the supported PDFium version from 8044 to 8057, for both the CGO and the WebAssembly backend. The bundled WebAssembly module has been rebuilt from 8057.

New API

• FPDFPageObj_GetRenderedStrokePattern (experimental): returns a bitmap rasterization of the tiling pattern a page object is stroked with.

Performance

Because patches CL 156070 and CL 156090 landed in this release, this gives a nice 6.3% speed improvement when we rendered a testset of 5000 PDFs. This speed improvement is only on the native version, the WebAssembly build already contained the patches since v1.20.0. If you care about rendering performance and use the native build, consider upgrading to PDFium 8057.

v1.20.2

What's Changed

• Resolve rect font information from the rect's own chars instead of a tolerance lookup by @​jerbob92 in klippa-app/go-pdfium#368
• Reload the page after FPDFPage_Flatten so the same page reference renders the flattened content by @​jerbob92 in klippa-app/go-pdfium#369
• Accept Windows device context handles in FPDF_RenderPage by @​jerbob92 in klippa-app/go-pdfium#370
• Rewrite some tests into ginkgo by @​jerbob92 in klippa-app/go-pdfium#371

Full Changelog: klippa-app/go-pdfium@v1.20.1...v1.20.2

Fixes

Rect font information no longer depends on a tolerance lookup (#368, fixes #23)

GetPageTextStructured with CollectFontInformation used FPDFText_GetCharIndexAtPos at the rect's top-left corner with a hardcoded tolerance of 5 points to find the rect's font. That lookup is wrong by construction: a rect is the union of glyph boxes, so its corner is not on the first char whenever a later char is taller, and the position lookup returns the first char in page order whose box contains the point, from any text object. Measured on the repo corpus, 11% of rects got zero-valued or wrong font information; on a set of 5000 real-world PDFs (1.38M rects), 0.5% did, across 1522 files. No tolerance value fixes it: larger values remove the not-found cases but increase the wrong-font cases.

The first char of each rect is now derived from the geometry in textextract (Extractor.FirstCharIndices), replaying how PDFium builds its rects. Text objects redrawn over each other are disambiguated with the next rect. Verified against an exact ground truth built from text object identity: zero wrong fonts on 1.38M rects. FontInformation is still never nil.

FPDFPage_Flatten reloads the page (#369, fixes #199)

PDFium parses a page's content when it is loaded and never again. Flattening rewrites the page dictionary, so rendering or extracting text from the same page handle showed the old content with the annotations gone. Only a freshly loaded page showed the flattened result, which is why "flatten ByReference, render ByIndex" happened to work.

After a flatten that returns Success, both backends reload the page behind the given reference. ByReference and ByIndex callers see the flattened content without code changes. Form fill environments that had the page loaded are moved to the reloaded page. The old page stays alive until the handle is closed, so text pages, page objects and annotations obtained before flattening remain valid memory (they keep pointing at the unflattened content, as documented on FPDFPage_Flatten).

FPDF_RenderPage accepts real Windows device context handles (#370, supersedes #270)

The Windows-only FPDF_RenderPage required request.DC to be a C.HDC, a cgo type private to the implementation package, so no caller outside it could ever satisfy the type check. It now accepts the handle as the Windows API hands it out: uintptr, syscall.Handle or unsafe.Pointer, with descriptive errors for other types and for a null handle.

The dllexport change proposed in #270 is not needed: Go 1.26 removed __declspec(dllexport)golang/go#56994

v1.20.1

What's Changed

• Bump github.com/onsi/ginkgo/v2 from 2.32.1 to 2.32.2 by @​dependabot[bot] in klippa-app/go-pdfium#363

... (truncated)

Commits

• 175fb06 Merge pull request #372 from klippa-app/feature/update-pdfium-to-8057
• db4291d Upgrade pdfium to 8057
• 458943e Merge pull request #371 from klippa-app/feature/ginkgo-tests
• 674950b Rewrite some tests into ginkgo
• 74bb3f4 Merge pull request #370 from klippa-app/feature/fix-windows-builds
• e4e4779 Merge pull request #369 from klippa-app/feature/fix-page-flattening
• 46b4341 Fix render tests for Windows
• de9a181 Skip form tests on multi
• 8624219 Fix rendering on Windows to a device
• c16b62d Fix page flattening
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore <dependency name> major version will close this group update PR and stop Dependabot creating any more for the specific dependency's major version (unless you unignore this specific dependency's major version or upgrade to it yourself)
• @dependabot ignore <dependency name> minor version will close this group update PR and stop Dependabot creating any more for the specific dependency's minor version (unless you unignore this specific dependency's minor version or upgrade to it yourself)
• @dependabot ignore <dependency name> will close this group update PR and stop Dependabot creating any more for the specific dependency (unless you unignore this specific dependency or upgrade to it yourself)
• @dependabot unignore <dependency name> will remove all of the ignore conditions of the specified dependency
• @dependabot unignore <dependency name> <ignore condition> will remove the ignore condition of the specified dependency and ignore conditions