feat(ppt): pictures from the BLIP store, real slide size, slide names#610
Open
andiwand wants to merge 3 commits into
Open
feat(ppt): pictures from the BLIP store, real slide size, slide names#610andiwand wants to merge 3 commits into
andiwand wants to merge 3 commits into
Conversation
Parse the StyleTextPropAtom ([MS-PPT] 2.9.44) that follows each text atom — in the drawing's ClientTextbox as well as the outline (SlideListWithText) — and split slide text into styled spans carrying bold, italic, underline, font size, font name (FontCollection) and explicit sRGB colors. Replaces the flat 11pt placeholder; unformatted text defaults to 18pt, approximating the unread master text styles. Paragraph-level runs are skipped field-by-field via their PFMasks; paragraphs store their first run's style so empty paragraphs keep their height. Scheme-indexed colors and master-inherited formatting remain open (documented in AGENTS.md). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01R9Z3kyUaHNkJacAEggevHH
Resolve each shape's pib (picture shape) or fillBlip (picture fill, the LibreOffice pattern) property through the OfficeArtBStoreContainer and the /Pictures delay stream, and emit JPEG/PNG BLIPs as image elements under the shape's frame, served as in-memory files. Background shapes (OfficeArtFSP.fBackground) get a full-slide anchor and render below the other shapes. The drawing's optional non-grouped shape ([MS-ODRAW] 2.2.13) is now read too - picture fills live there. Also: slide size from the DocumentAtom instead of hardcoded 10x7.5in, and slide_name returns "Slide N". WMF/EMF/PICT/DIB/TIFF BLIPs and pictures on master slides remain open (documented in AGENTS.md). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01R9Z3kyUaHNkJacAEggevHH
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 43863a237f
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
Comment on lines
+257
to
+259
| [[nodiscard]] std::string image_href( | ||
| [[maybe_unused]] const ElementIdentifier element_id) const override { | ||
| return {}; |
There was a problem hiding this comment.
When callers set HtmlConfig::embed_images = false, the generic image renderer asks the resource locator for a URL based on image.href(). Returning an empty string here makes every PPT image resource use an empty path, so the generated HTML gets src="" and bring_offline() cannot write distinct image assets. This only shows up in non-embedded image output, but that is a supported renderer configuration.
Useful? React with 👍 / 👎.
With HtmlConfig::embed_images = false the generic renderer uses
image.href() as the resource path, so the empty href produced src=""
and bring_offline could not write distinct image assets. Pictures have
no container path in the file; synthesize one from the one-based BLIP
store index plus the extension matching the image bytes
("Pictures/<index>.png|.jpg"). Addresses review feedback on #610.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AEvWWwe1EyeWH7ANHoAvPH
andiwand
force-pushed
the
feat/ppt-character-formatting
branch
from
July 22, 2026 05:53
7a2bd37 to
97ebb8e
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
🤖 Generated with Claude Code
Stacked on #609 (same files; base is
feat/ppt-character-formatting).Adds pictures to the legacy
.pptdecoder, plus two small fidelity wins from the AGENTS.md open-work list.Pictures
OfficeArtFOPTis now parsed forpib(real picture shapes) andfillBlip(picture fills — how LibreOffice-exported.pptplaces pictures); both are one-based indexes into theOfficeArtBStoreContainerof the document's drawing group.OfficeArtFBSElocates its BLIP in the/Pictures(delay) stream atfoDelay(embedded BLIPs are handled too). JPEG and PNG BLIPs become animageelement under the shape's frame, served to the renderer as an in-memoryodr::File→ rendered as a base64<img>. WMF/EMF/PICT/DIB/TIFF are skipped (documented open work).fBackground(OfficeArtFSP) get a full-slide anchor when they carry none and are ordered before the slide's other shapes so they render underneath.Micro-wins
DocumentAtom(master units → inches) instead of the hardcoded 10"×7.5" —style-various-1.pptrenders at its real 11.02"×8.27" size (open-work §2.1).slide_namereturns "Slide N" in presentation order (open-work §2.6).Validation & tests
OldMs.ppt_style_variousextended: slide 6 ("title6 – background image") now asserts a leading full-slide frame holding an image element whose bytes are the PNG from/Pictures; frame count 12 → 13. AllOldMs.*and pptHtmlOutputTestspass.Note: the
2025-09-11photo-deck fixture keeps rendering without photos — its pictures live on per-slide masters, which are skipped entirely (documented open work).https://claude.ai/code/session_01R9Z3kyUaHNkJacAEggevHH