added xlsx reader #11287
added xlsx reader #11287
Conversation
aantich
commented
Nov 9, 2025
aantich
commented
- Reads multiple spreadsheets
- Converts tables to tables
- Basic tests added
.gitignore
Outdated
| /test-docs | ||
| doc/pptx-reader-design-v2.md | ||
| doc/pptx-reader-design.md | ||
| doc/xlsx-reader-design.md |
There was a problem hiding this comment.
As with the other PR, let's leave off these .gitignore changes.
|
I think the test files may need to be included in extra-source-files in pandoc.cabal. |
|
The new format should be added to MANUAL.txt ; see the list under |
|
One thing I noticed testing this locally: I got a table with hundreds of empty rows. |
We'll fix the trailing empty rows; for empty rows in the middle of the sheet I believe it makes sense to keep them, since I may want to reference the exact cell coordinates eventually, e.g. a table like this: is converted to markdown like this: Main {#sheet-1}Person Age Location Anton Antich 23.0 Switzerland Just a random cell I believe this is the correct behavior. |
- MANUAL updated - trailing empty rows removed
|
Believe we addressed all comments. Didnt test empty rows rigorously, but on a couple of quick files it works. |
| cellToInlines :: XlsxCell -> [Inline] | ||
| cellToInlines cell = | ||
| let base = case cellValue cell of | ||
| TextValue t -> [Str t] |
There was a problem hiding this comment.
For text values, better to use B.toList (B.text t) where B is Text.Pandoc.Builder. This will convert the string into a list of Str and Space elements. Pandoc expects spaces to be represented as Space and not space characters inside a Str.
| nativeDiff :: FilePath -> Pandoc -> Pandoc -> IO (Maybe String) | ||
| nativeDiff normPath expectedNative actualNative | ||
| | expectedNative == actualNative = return Nothing | ||
| | otherwise = Just <$> do | ||
| expected <- T.unpack <$> runIOorExplode (writeNative def expectedNative) | ||
| actual <- T.unpack <$> runIOorExplode (writeNative def actualNative) | ||
| let dash = replicate 72 '-' | ||
| let diff = getDiff (lines actual) (lines expected) | ||
| return $ '\n' : dash ++ | ||
| "\n--- " ++ normPath ++ | ||
| "\n+++ " ++ "test" ++ "\n" ++ | ||
| showDiff (1,1) diff ++ dash |
There was a problem hiding this comment.
Since this seems to be duplicated from docx reader tests, I wonder if it makes sense to import it from there, or put it in some common place, e.g. Test.Helpers ?
2 participants
