Does PyMuPDF4LLM keep a table together across a page break?
PyMuPDF4LLM produced two Markdown tables for each document, preserving the repeated-header version but mishandling the headerless continuation. It kept 22 of 22 data rows in the first; in the second, a data row became headings, leaving 25 of 26 data rows and incorrect continuation column names.
1 of 2 test cases passed · 1 failed
Every test case in this scenario has a result.
The 2 test cases
Each test case is judged on its own: Pass, Fail, Not gradable, or Untested. The scenario result above is based on these test cases.
Cross-page table — one table, completePassEvidence
We converted the monitoring-station PDF to Markdown using the library's standard conversion entry point, with no optional arguments, and saved the result.
Copied from Output file (Markdown), lines 89–108, 112–116, 118–123
Braithe Estuary Partnership · Water quality programme # **Monitoring stations: schedule and sampling intervals** Water quality programme · Schedule current at 1 February 2026 The schedule below lists the fixed stations sampled under the water quality programme. Stations are visited in the order shown wherever the tide allows. Depths are metres below lowest astronomical tide as recorded at installation, and are not corrected for later change. ## **The programme** The water quality programme has run in its present form since 2011, when the first two stations were established at Vardenne Quay and Thrimby Point. There are now twenty-two, the most recent added in November 2025. The programme provides the baseline against which any change at Kellow Sands is measured, and it is the only continuous record of salinity in the upper estuary. Sampling is carried out by the Partnership's own staff, with the Vellamy Society providing cover during August. Analysis is carried out under contract by Kirkhollow Analytical Laboratories. The determinations are those set by the 2015 programme review: suspended solids, chloride, ammoniacal nitrogen, oxidised nitrogen, orthophosphate and dissolved organic carbon, with particle size on the intertidal stations twice a year. ## **Sampling** Samples are taken from a boat at the deep stations and by wading at the intertidal ones, in each case on the ebb and within two hours of high water. Where a station cannot be reached on the day it is due it is taken as soon afterwards as conditions allow, and the actual date is recorded: the schedule is a plan and not a record. Stations on a weekly interval are the exception, and a missed week there is reported to the Board. Field measurements of temperature and dissolved oxygen are taken at every visit, entered in the field book, and transcribed into the station record within the week. Bottles are delivered to Braithe Road the same day wherever the round permits, and are held in the refrigerated drop box overnight when it does not. ## **The stations** Grid references are to the nearest ten metres and describe the position of the marker, not of the sampling point, which at the intertidal stations is up to five metres seaward of it. Two stations are sampled quarterly rather than monthly; both were added to answer a particular question about the southern creeks, and both are due for review in 2027. Access is by public footpath at every station except Marnhead Deep and Tarbert Jetty, both of which need the harbour master's permission and a crew of two. The Partnership's insurance covers wading only where a second person is on the shore, and staff are asked not to work the intertidal stations single-handed even where the access itself is straightforward. ## **Reporting** Results are entered in the station record within a week of the report arriving from the laboratory, and the record is closed for the quarter at the end of the following month. A late result is entered with the date it arrived as well as the date of sampling, because the two have diverged by as much as five weeks where the laboratory has had to repeat a batch. The quarterly summary goes to the Board and to the Harbour Board, and an annual summary is published. Neither includes the field measurements, which are held in the station record and are available to anyone who asks the secretariat for them. Page 1 of 4 Braithe Estuary Partnership · Water quality programme ## **Changes to the schedule** A station is added or withdrawn only by a decision of the Board, on a recommendation from the secretariat, and the reason is minuted. The programme has lost three stations since 2011: two at the head of Kellow Creek, which dried out after the storms of 2014, and one at the bar, which could not be reached safely often enough to be worth keeping. A change to a sampling interval does not need a Board decision and is made by the secretariat where the tide or the staffing requires it. Every such change is listed in the quarterly summary. Two stations moved from monthly to quarterly in 2024 for that reason and both are due to be reviewed. ## **Equipment** Field measurements are taken with two hand-held meters, both serviced annually under contract and both calibrated against the same standards before each round. The meters are identified in the station record by serial number, because a drift on one of them in 2022 was traced only because the record said which meter had been used. The boat is the Partnership's own and is kept at the boatyard. It carries no sampling equipment of its own: everything goes aboard for the round and comes off afterwards, which is slower but has stopped bottles being left in the sun. ## **Health and safety** Every station has a written access note, and the notes are reviewed each January. Two stations need a permit and a crew of two; three are on soft ground where a wading pole is carried; and one, at the training wall, is worked only on a falling tide and never alone. The notes are in the boat's folder and in the office copy of this schedule. Staff working the intertidal stations carry a radio and tell the office which round they intend to work and when they expect to be off the flats. The office rings the duty officer if a party has not reported in within an hour of that time. The arrangement has been tested twice and worked on both occasions. ## **The record** The station record is a spreadsheet held on the Partnership's own drive, with a copy taken monthly and held by the Harbour Board. It carries one row per visit: the station, the date, the time, the state of the tide, the field measurements, the laboratory reference and the results when they arrive. Nothing is overwritten. A corrected value is entered as a new row with a note, and the original stays where it was. The record goes back to March 2011 for the original stations and to the month of installation for the rest. The first two years are less complete than the Partnership would like: the state of the tide was not recorded at all until 2013, and there is a gap of eleven weeks in the winter of 2012 when the meter was away for repair. ## **The 2015 review** The programme was reviewed in 2015 by a panel of three, one of them from outside the estuary, and the review is worth reading before any change is proposed. Its main recommendation, that the number of determinations be cut and the number of visits kept, was accepted; the programme has been cheaper and more useful since. Its second recommendation, that the Partnership publish the data as it is collected rather than annually, has not been implemented and is raised at most Board meetings. The panel also recommended that two stations be moved: one at the head of Kellow Creek, which was moved forty metres seaward in 2016, and one at the bar, which was withdrawn instead. The reasons for withdrawing rather than moving it are in the minute of that meeting and are still sound. Page 2 of 4 Braithe Estuary Partnership · Water quality programme ## **Access to the data** The record is available to Partnership members on request and to anyone else at the secretariat's discretion, which in practice means that a request from a school, a consultant or a neighbour is met, and that a request with no reason given is asked for one. There is no charge for a copy of the record; there is a charge for the secretariat's time in extracting something particular from it. The annual summary is published in April and is sent to the parish councils, the wildlife trust and the two angling clubs. It carries the means and the ranges by station and by quarter rather than the individual results, which is a decision of the Board and not a limitation of the record. ## **Costs** The programme costs about nineteen thousand pounds a year, of which the laboratory's charges are rather more than half and the rest is staff time, the boat and the meters. The Partnership meets two thirds and the Harbour Board one third under the agreement of 2015, which runs to 2027 and is due to be renegotiated in the coming year. ## **Determinations** Every visit produces a sample for suspended solids and chloride. Ammoniacal nitrogen and oxidised nitrogen are taken at the twelve stations above the training wall, orthophosphate at the six in the creeks, and dissolved organic carbon at four stations chosen in 2015 to give a spread across the estuary rather than a concentration in one reach. Particle size is taken twice a year at the intertidal stations and is the one determination the programme has kept against advice. The 2015 panel thought it of little value at this frequency and the Partnership has continued it because the intertidal survey needs a comparison, which is a reason the panel did not have in front of it. Table 1. Fixed monitoring stations, with grid references and sampling intervals. |**Station**|**Grid reference**|**Depth (m)**|**Sampling interval**|**Installed**| |---|---|---|---|---| |Vardenne Quay|NX 4118 6602|2.4|Fortnightly|March 2011| |Thrimby Point|NX 4276 6491|1.8|Fortnightly|March 2011| |Kellow Sands north|NX 4390 6588|0.9|Monthly|June 2013| |Kellow Sands south|NX 4402 6531|1.1|Monthly|June 2013| |Marnhead Deep|NX 4025 6714|8.6|Monthly|October 2014| |Braithe Bridge|NX 3982 6853|3.2|Weekly|April 2015| |Harrowby Reach|NX 4147 6770|4.5|Fortnightly|April 2015| |Sandiman Bank|NX 4233 6685|1.3|Monthly|September 2016| |Vellamy Creek|NX 4318 6438|0.6|Monthly|September 2016| |Tarbert Jetty|NX 4064 6552|2.9|Fortnightly|May 2018| |Croyde Gut|NX 4351 6367|1.6|Quarterly|May 2018| |Prosser Flats|NX 4189 6304|0.7|Quarterly|August 2020| |Old Lifeboat House|NX 4295 6620|1.9|Fortnightly|August 2020| |Training Wall east|NX 4126 6459|5.4|Weekly|February 2023| |Lissamer Point|NX 4208 6742|3.7|Monthly|February 2023| |Kellow Creek head|NX 4436 6612|0.4|Quarterly|March 2024| |Vardenne Outfall|NX 4102 6571|2.1|Weekly|March 2024| |Marnhead Bar|NX 3968 6741|9.3|Monthly|September 2024| Page 3 of 4 Braithe Estuary Partnership · Water quality programme ### Table 1 (continued) |**Station**|**Grid reference**|**Depth (m)**|**Sampling interval**|**Installed**| |---|---|---|---|---| |Thrimby Spit|NX 4301 6467|1.4|Fortnightly|September 2024| |Harrowby Cut|NX 4163 6799|5.8|Monthly|April 2025| |Boatyard Steps|NX 4088 6588|1.2|Fortnightly|April 2025| |Cavenshaw Bight|NX 4372 6521|2.6|Quarterly|November 2025| Croyde Gut and Prosser Flats are sampled in the first week of February, May, August and November. Stations withdrawn before 2011 are listed in the programme review of that year. A station is sampled on the interval shown or better. Where a visit is missed the record says so and the next visit is not brought forward to compensate, because a shortened interval is harder to allow for later than a missing week. Page 4 of 4
Returned Markdown: 22 station rows split between two tables
Open the Markdown file ↗| ✓ | Expected: Retain all 22 data rows.Found: The output contains 22 data rows, matching the PDF's 22: 18 in the first table and 4 in the second. |
| ✓ | Expected: Preserve Thrimby Spit's grid reference, depth, sampling interval and installation date.Found: Thrimby Spit retains grid reference NX 4301 6467, depth 1.4 m, fortnightly sampling and installation in September 2024. |
| ✓ | Expected: Preserve Harrowby Cut's grid reference, depth, sampling interval and installation date.Found: Harrowby Cut retains grid reference NX 4163 6799, depth 5.8 m, monthly sampling and installation in April 2025. |
| ✓ | Expected: Preserve Boatyard Steps' grid reference, depth, sampling interval and installation date.Found: Boatyard Steps retains grid reference NX 4088 6588, depth 1.2 m, fortnightly sampling and installation in April 2025. |
| ✓ | Expected: Preserve Cavenshaw Bight's grid reference, depth, sampling interval and installation date.Found: Cavenshaw Bight retains grid reference NX 4372 6521, depth 2.6 m, quarterly sampling and installation in November 2025. |
| ✓ | Expected: Keep Marnhead Bar before Thrimby Spit across the page break.Found: Marnhead Bar ends the first table; Thrimby Spit starts the second. |
| ✓ | Must not appear: Do not treat repeated column headings as data; a second table with those headings is allowed.Found: The repeated headings became the second table's header row, not a data row. |
| ✓ | Must not appear: Keep page numbers, the running header and the continuation caption outside table bodies.Found: Page numbers, the running header and the continuation caption remain outside both table bodies. |
Two Markdown tables were allowed if the continuation retained its headers. The output preserved the required rows and column relationships, so the split did not prevent a pass.
Continuation caption formatting. The continuation caption "Table 1 (continued)" became a third-level Markdown heading rather than plain text.
Cross-page table — continuation without a repeated headerFailEvidence
We converted the sampling-visits PDF to Markdown using the library's standard conversion entry point, with no optional arguments, and saved the result.
Copied from Output file (Markdown), lines 17–19, 32–52
Page 1 of 2 Braithe Estuary Partnership · Water quality programme # Sampling visits, winter 2026 Water quality programme · Visits between 6 January and 19 February 2026 Table 2 lists every sampling visit made in the first seven weeks of 2026. A visit is one boat or wading call at one station on one morning; where a station is visited twice in the period it appears twice. Determinations are those taken on the visit, not those scheduled. Times are the time the sample was drawn. Method. Each visit follows the written access note for the station. Field measurements of temperature and dissolved oxygen are taken first and entered in the field book; the bottle is then filled below the surface, labelled with the station code and the time, and placed in the cool box. Where the tide prevents a wading station being reached within two hours of high water the visit is recorded as missed and the next scheduled visit is not brought forward. Samples are delivered to Braithe Road the same day and held in the refrigerated drop box overnight where the round permits. Record. The table below is transcribed from the field books for the period and checked against the laboratory reception log. A visit appears once per station per morning. Determinations are those actually taken, which may differ from the schedule where a meter was out of calibration or a bottle was lost; the two such cases in this period are noted in the quarterly summary and not in this table. Table 2. Sampling visits, 6 January to 19 February 2026. |Visit|Station|Date|Time|Determinations taken|Sampler| |---|---|---|---|---|---| |V-01|Vardenne Quay|2026-01-06|09:40|Suspended solids, chloride|J. Marren| |V-02|Thrimby Point|2026-01-06|10:25|Suspended solids, chloride|J. Marren| |V-03|Kellow Sands north|2026-01-08|08:55|Suspended solids, chloride, orthophosphate|R. Ashby| |V-04|Kellow Sands south|2026-01-08|09:30|Suspended solids, chloride, orthophosphate|R. Ashby| |V-05|Marnhead Deep|2026-01-09|11:10|Suspended solids, chloride, DOC|J. Marren| |V-06|Braithe Bridge|2026-01-12|07:50|Suspended solids, chloride, nitrogen|P. Sowerby| |V-07|Harrowby Reach|2026-01-12|08:40|Suspended solids, chloride, nitrogen|P. Sowerby| |V-08|Sandiman Bank|2026-01-13|10:05|Suspended solids, chloride|R. Ashby| |V-09|Vellamy Creek|2026-01-13|10:50|Suspended solids, chloride, orthophosphate|R. Ashby| |V-10|Tarbert Jetty|2026-01-14|09:15|Suspended solids, chloride|J. Marren| |V-11|Croyde Gut|2026-02-03|08:20|Suspended solids, chloride, orthophosphate|P. Sowerby| |V-12|Prosser Flats|2026-02-03|09:05|Suspended solids, chloride, orthophosphate|P. Sowerby| |V-13|Old Lifeboat House|2026-02-04|10:30|Suspended solids, chloride|R. Ashby| |V-14|Training Wall east|2026-02-05|07:45|Suspended solids, chloride, nitrogen|J. Marren| Page 2 of 2 Braithe Estuary Partnership · Water quality programme |V-15|Lissamer Point|2026-02-05|08:35|Suspended solids, chloride, nitrogen|J. Marren| |---|---|---|---|---|---| |V-16|Kellow Creek head|2026-02-06|11:20|Suspended solids, chloride, orthophosphate|R. Ashby| |V-17|Vardenne Outfall|2026-02-09|07:30|Suspended solids, chloride, DOC|P. Sowerby| |V-18|Marnhead Bar|2026-02-09|08:50|Suspended solids, chloride, DOC|P. Sowerby| |V-19|Thrimby Spit|2026-02-10|09:25|Suspended solids, chloride|J. Marren| |V-20|Harrowby Cut|2026-02-11|10:15|Suspended solids, chloride, nitrogen|R. Ashby| |V-21|Boatyard Steps|2026-02-11|11:00|Suspended solids, chloride|R. Ashby| |V-22|Cavenshaw Bight|2026-02-12|08:10|Suspended solids, chloride, orthophosphate|P. Sowerby| |V-23|Braithe Bridge|2026-02-16|07:55|Suspended solids, chloride, nitrogen|J. Marren| |V-24|Vardenne Outfall|2026-02-16|08:40|Suspended solids, chloride, DOC|J. Marren| |V-25|Training Wall east|2026-02-19|07:50|Suspended solids, chloride, nitrogen|P. Sowerby| |V-26|Harrowby Reach|2026-02-19|08:45|Suspended solids, chloride, nitrogen|P. Sowerby| Visits V-11 and V-12 were made on the first Tuesday of February under the quarterly arrangement for the two creek stations. No visit was missed in the period.
Returned Markdown: a visit used as headings and 25 remaining data rows
Open the Markdown file ↗| ✗ | Expected: Retain all 26 data rows.Found: The output has 25 data rows against 26 in the PDF: 14 in the first table and 11 in the second. |
| ✗ | Expected: Keep visit 15 as a data row with its original values.Found: Visit 15's values became the second table's headings; no data row remains for that visit. |
| ✓ | Expected: Preserve visit 22's station, date, time, measurements and sampler.Found: Visit 22 retains Cavenshaw Bight, 12 February 2026, 08:10, suspended solids, chloride, orthophosphate and its original sampler, under incorrect headings. |
| ✓ | Expected: Preserve visit 26's station, date, time, measurements and sampler.Found: Visit 26 retains Harrowby Reach, 19 February 2026, 08:45, suspended solids, chloride, nitrogen and its original sampler, under incorrect headings. |
| ✓ | Expected: Keep visit 14 before visit 15 across the page break.Found: Visit 14 ends the first table; visit 15's content appears next, as the second table's headings. |
| ✗ | Expected: Keep continuation rows under the original six column headings, in one table or a second table.Found: Visits 16 through 26 appear under headings taken from visit 15's values, rather than the original six column headings. |
| ✓ | Must not appear: Keep page numbers and the running header outside table bodies.Found: The page number and running header sit between the two tables, outside both table bodies. |
The continuation needed its original columns. The default conversion completed but turned its first row into headings, so this was a table-extraction failure rather than a blocked conversion. This result concerns the default pipeline, not every available table-detection option.
Remaining visit values. Visits 16 through 26 retained their original cell values.
Configuration and setup
How this tool was set up for the run and what the test needed in place. Each row is a fact from the run's records; a fact the records do not hold is left out, not guessed.
How this scenario is graded
How we decide Pass, Fail and Not gradable. The same rules apply to every tool tested on this scenario.
Each test case gets one result per tool: Pass, Fail or Not gradable. A test case we haven't run yet shows Untested. There are no partial results.
- Pass: the tool did everything the test expected, and nothing it said contradicts the correct answer.
- Fail: at least one expected behaviour clearly didn't happen; the test case's result says which one and quotes the tool.
- Not gradable: our evidence couldn't settle the outcome (for example a record we needed is missing). It is never counted as a fail, and the test case's result says what's missing.
Where this sits in the benchmark
This page is one cell of a larger study: one tool, one scenario. Only this benchmark's frame appears here.
| Level | Name | Scope |
|---|---|---|
| Benchmark | Converting a complex PDF into clean Markdown with an open-source library → | 12 scenarios · 10 tools |
| Capability | Table Extraction → | |
| Scenario | A table that continues across a page break → | |
| Tool | PyMuPDF4LLM → |
Global scenario definition → · Global capability definition → · PyMuPDF4LLM product page →
History of this result
What has happened to this result since it was first published. Runs and grades are never overwritten: a retest or a re-grade publishes a new result and keeps the earlier one readable.
Act on this result
Nothing filed here edits the run or the grade. A challenge opens a review, and a review can produce a new run or a re-grade — which becomes the current result and leaves this one in the history.
You run the same kind of test against your own setup and get the same behaviour.
Agree →Yours behaves differently. Tell us what you got, with a screenshot if you have one.
Disagree →Something here is wrong — a reference value, a transcription, a grade.
Report an issue →Tell us what changed and we schedule a rerun of the failing test case. The old result stays as history.
Vendor notice →The same record is available as structured data through the AI Demos MCP server, with the counts, the coverage and every per-test-case reason carried as fields.
Verify the proof files
These files support this result. Open a file to inspect the original evidence.
File fingerprints (SHA-256)
A fingerprint identifies the exact file used for this result.