How to consolidate paper links into a reference block
- Step 1Open the tidy tool — Go to /markdown-tools/md-footnote-linker. The paper is processed in your browser and never uploaded.
- Step 2Load the paper — Paste the Markdown or drop one
.mdfile (one document per run — acceptsMultiple is false). - Step 3Run the tidy — Click Run. No options exist — the consolidation is a single deterministic pass.
- Step 4Verify the body — Inline
[text](https://url)citations are now[text][n]; duplicates merged; footnotes and code unchanged. - Step 5Inspect the reference block — The appended
[n]: urlblock lists each unique source once, in citation order — your tidy reference list. - Step 6Convert and submit — Copy the cleaned paper, then convert with /markdown-tools/md-to-docx (or Pandoc) for the journal's required format.
Pre-submission link pass: what happens to each construct
Only inline http(s) links are consolidated. Scholarly apparatus is preserved so footnote and citation checking stays in your hands.
| Construct | Action | Submission note |
|---|---|---|
[Author Year](https://doi.org/...) | Consolidated → [Author Year][n] | DOI citations become numbered references. |
[dataset](https://zenodo.org/...) | Consolidated → [dataset][n] | Any http(s) resource link is collected. |
[note](https://x.org "caption") | Consolidated; title dropped | Captions in titles are removed from the block. |
[^1] + [^1]: citation | Preserved | Real footnotes are NOT validated or renumbered. |
[Methods](#methods) | Preserved | Internal anchors stay inline. |
[supplement](supp.md) | Preserved | Relative file links untouched. |
| Citation example in a ``` block | Preserved | Code fences are excluded from the pass. |
Limits for paper-length documents
A full paper can hit the character cap before the size cap. The character cap is enforced independently of bytes.
| Plan | Max file size | Max characters | Files per run |
|---|---|---|---|
| Free | 1 MB | 500,000 | 1 |
| Pro | 10 MB | 5,000,000 | 10 |
| Pro-media | 50 MB | 20,000,000 | 50 |
| Developer | 500 MB | Unlimited | Unlimited |
Cookbook
Before/after pairs from research papers, framed as a pre-submission tidy. DOIs and URLs are illustrative.
DOI links consolidated for submission
Inline DOIs become numbered references; the block is your tidy source list ready for the journal.
Before: We build on [Lakatos (1976)](https://doi.org/10.1/l) and [Kuhn (1962)](https://doi.org/10.2/k). After: We build on [Lakatos (1976)][1] and [Kuhn (1962)][2]. [1]: https://doi.org/10.1/l [2]: https://doi.org/10.2/k
Key paper cited once in references
A foundational source cited five times collapses to a single block entry — clean for submission.
Before: [Kuhn (1962)](https://doi.org/10.2/k) ... [Kuhn again](https://doi.org/10.2/k). After: [Kuhn (1962)][1] ... [Kuhn again][1]. [1]: https://doi.org/10.2/k
Footnotes need separate checking
The tool leaves [^id] footnotes untouched — it will NOT catch an orphan, so footnote QA stays a separate step.
Before: A caveat.[^2] See [data](https://data.org/d). (No [^2]: definition exists — an orphan.) After: A caveat.[^2] See [data][1]. (The orphan [^2] is unchanged — not detected.) [1]: https://data.org/d
Supplementary relative links preserved
Links to supplementary files stay inline, so the paper's internal structure is intact after the tidy.
Before: See [supplement](supp.md) and the external [code](https://github.org/x). After: See [supplement](supp.md) and the external [code][1]. [1]: https://github.org/x
Convert to .docx after tidying
Run the link tidy first, then convert — never the other way around, since fixing source after conversion is painful.
Workflow: 1. md-footnote-linker -> consolidate inline links 2. md-to-docx -> Word file for the journal Why this order: editing reference-style Markdown is trivial; editing a generated .docx is not.
Edge cases and what actually happens
Orphaned footnotes are not detected
Not supportedThe tool does not find footnote references without definitions, definitions without references, or duplicate [^id] IDs. It consolidates inline [text](url) links only. Do footnote QA separately.
Citation accuracy is not verified
Out of scopeNo checking of citation content, CSL, biblatex, or endnote/footnote semantics. The tool rewrites inline links to reference-style notation and nothing more.
Output renders as hyperlinks
By designReference-style links ([text][n] ... [n]: url) render as ordinary inline hyperlinks, not as [^n] superscript footnotes with a footnotes section.
Link captions (titles) are dropped
DroppedAn optional "title" after a URL is matched and removed; the block holds bare URLs only.
Only http(s) links are consolidated
PreservedRelative supplement links and in-page anchors are not matched and stay inline. Only http:// and https:// URLs enter the block.
Repeated source
DeduplicatedAll occurrences of one URL share a number; the block lists it once — exactly what a clean reference list needs.
URL with whitespace
Not matchedThe URL pattern stops at whitespace, so an unencoded space prevents the match and the link stays inline. Percent-encode (%20) first.
No inline links present
UnchangedA paper using only [^id] footnotes and relative links returns identical output — nothing collected, no block appended.
Paper exceeds the character cap
RejectedFull papers can exceed the Free 500,000-character cap under 1 MB on disk. The cap is independent of file size; upgrade or split with /markdown-tools/md-splitter.
Frequently asked questions
Is this a footnote cleanup that finds orphans and duplicates?
No. It does not detect orphaned footnotes, missing definitions, or duplicate [^id] IDs, and it does not validate citations. It consolidates inline [text](url) links into a reference-style block. Footnote QA is a separate step.
Should I run it before or after PDF/.docx generation?
Before. Editing reference-style Markdown is trivial; editing a generated .docx or PDF is not. Tidy links first, then convert with /markdown-tools/md-to-docx.
Does it integrate with Pandoc?
Yes. The output is valid CommonMark, so Pandoc parses the reference-style links as normal hyperlinks during conversion to .docx or PDF.
What about endnotes vs. footnotes?
This tool does not produce or manage either — it makes reference-style links that render as hyperlinks. Footnote/endnote handling depends on your renderer and is outside this tool's scope.
Does it touch my [^id] footnotes?
No. Real footnotes and their definitions are preserved verbatim; they are not links and are never matched or validated.
Are duplicate citations merged?
Yes. Repeated occurrences of one URL share a number, and the reference block lists each unique source once, in citation order.
Are link captions kept?
No. The optional "title" is dropped. If you need to keep titles, use /markdown-tools/md-ref-link-converter, which preserves them.
Are there options to configure?
No. It is a single deterministic pass with no UI controls.
Is my unpublished paper uploaded?
No. The transform runs in your browser; nothing is sent to a server.
How do I process a multi-section paper?
Process one file per run, or merge sections first with /markdown-tools/md-merger for a single reference block — watch the character cap on Free.
Can I confirm the cited URLs resolve?
Run the output through /markdown-tools/md-link-validator to flag malformed URLs, relative paths, and DNS failures (it cannot detect HTTP 404 from the browser).
What is the final pre-submission sequence?
Tidy links here, optionally normalize with /markdown-tools/md-prettifier, then convert with /markdown-tools/md-to-docx for the journal's required format.
Privacy first
All Markdown processing runs locally in your browser using JavaScript. No file is ever uploaded to JAD Apps servers — only metadata counters are saved for signed-in dashboard stats.