What causes "slide part is missing: {resolved}" in nexu-io/open-design?

List the archive contents to see which slide parts exist: `unzip -l deck.pptx | grep slides/slide` and compare against the rels targets.

nexu-io/open-design · error · ValueError

slide part is missing

Error message

slide part is missing: {resolved}

What it means

Raised in _ordered_slide_parts (pptx_qa.py) immediately after the relationship check: once a slide relationship Target is resolved into a part path, the code confirms that path exists in the .pptx zip via `zf.namelist()`. If the slide XML part referenced by the relationship is absent from the archive, it raises ValueError. This catches packages whose relationships point at parts that were never packaged.

Solutions

List the archive contents to see which slide parts exist: `unzip -l deck.pptx | grep slides/slide` and compare against the rels targets.
Round-trip the file through PowerPoint/Keynote to reconcile presentation.xml with the packaged parts, then retry.
Regenerate the deck from the upstream producer (python-pptx script, exporter) rather than patching the broken zip by hand.
If only a few slides are missing, accept that the deck is damaged and run pptx_qa on a known-good backup copy.

Example fix

// before
python3 pptx_qa.py inspect deck.pptx
# -> ValueError: slide part is missing: ppt/slides/slide5.xml

// after
unzip -l deck.pptx | grep slide   # confirm which parts exist
# regenerate/re-save deck, then:
python3 pptx_qa.py inspect deck.pptx

Defensive patterns

Strategy: try-catch

Validate before calling

import zipfile

def all_slide_parts_present(path: str) -> bool:
    with zipfile.ZipFile(path) as zf:
        names = set(zf.namelist())
        try:
            from pptx_qa import _ordered_slide_parts  # reuse the resolver
            for part in _ordered_slide_parts(zf):
                if part not in names:
                    return False
        except ValueError:
            return False
    return True

Try / catch

try:
    parts = _ordered_slide_parts(zf)
except ValueError as exc:
    raise SystemExit(f"Deck is missing slide parts: {exc}. Re-export from the source tool.") from exc

Prevention

Confirm slide parts exist in the archive before parsing: `unzip -l deck.pptx | grep slides/slide`.
Re-export decks from the original tool rather than patching the zip.
When accepting user uploads, validate package integrity up front and reject corrupt files with a clear message.
Cross-check rels Targets against namelist() to catch dangling references early.

When it happens

Trigger: Running pptx_qa on a .pptx where presentation.xml.rels has a Target like 'slides/slide5.xml' but the archive contains no such entry; or where Target resolved through _resolve_target lands on a path not in namelist(). Common with stripped/slimmed packages or partially extracted decks.

Common situations: A tool that rewrote relationships without re-zipping the corresponding slide XML; file extraction/repackaging that dropped slide parts; a deck that was edited to remove slides but whose presentation.xml still lists them; cross-package Target resolution bugs in converters producing absolute or wrongly-relative paths.

Related errors

AI-assisted analysis of nexu-io/open-design@5be4028344 (2026-08-12). Data as JSON: /api/errors/68f10eddaec58809. Report an issue: GitHub.

Appendix: source

Thrown at plugins/community/humanize-ppt/scripts/pptx_qa.py:89

    if target.startswith("/"):
        return target.lstrip("/")
    return posixpath.normpath(posixpath.join(posixpath.dirname(source_part), target))


def _ordered_slide_parts(zf: zipfile.ZipFile) -> list[str]:
    presentation = "ppt/presentation.xml"
    root = _read_xml(zf, presentation)
    rels = _relationships(zf, presentation)
    parts: list[str] = []
    for slide_id in root.findall(".//p:sldIdLst/p:sldId", NS):
        rel_id = slide_id.attrib.get(f"{{{REL}}}id")
        rel = rels.get(rel_id or "", {})
        target = rel.get("Target")
        if not target:
            raise ValueError(f"slide relationship is missing for {rel_id or 'unknown id'}")
        resolved = _resolve_target(presentation, target)
        if resolved not in zf.namelist():
            raise ValueError(f"slide part is missing: {resolved}")
        parts.append(resolved)
    return parts


def _text(root: ET.Element) -> str:
    return " ".join(
        (node.text or "").strip()
        for node in root.findall(".//a:t", NS)
        if (node.text or "").strip()
    )


def _tokens(value: str) -> set[str]:
    return {
        token.lower()
        for token in re.findall(r"[A-Za-z0-9]+|[\u3400-\u9fff]", value)
        if token.strip()
    }

View on GitHub (pinned to 5be4028344)