Fix duplicated IDs in ALTO XML when multiple pages are present #4386
Conversation
jankal
changed the title
jankal
changed the title
|
Thanks for the fix. I just wonder whether we should try to keep the current IDs for the most common use case where only a single page is processed. This could be achieved for example by
|
…nt pge This will ensure, validated ALTO XML output is generated while keeping IDs for the first page consistent as before.
|
Thanks for your suggestion @stweil! I added a function Disclosure: I haven't written CPP before, so pls check for rookie mistakes :) |
There was a problem hiding this comment.
Pull Request Overview
This PR addresses the issue of duplicated IDs in ALTO XML output when processing multiple pages by incorporating the page number into the generated IDs.
- Introduces a new helper function (GetID) to generate IDs based on a prefix, page number, and counter.
- Updates ID assignments for elements such as Illustration, GraphicalElement, ComposedBlock, TextBlock, TextLine, and String to ensure uniqueness.
Comments suppressed due to low confidence (1)
src/api/altorenderer.cpp:54
- [nitpick] Consider renaming GetID to 'GenerateUniqueID' to more clearly reflect its role in creating unique IDs across pages.
static std::string GetID(char const * prefix, int page_number, int counter) {
| @@ -51,6 +51,20 @@ static void AddBoxToAlto(const ResultIterator *it, PageIteratorLevel level, | |||
| } | |||
| } | |||
|
|
|||
| static std::string GetID(char const * prefix, int page_number, int counter) { | |||
There was a problem hiding this comment.
| static std::string GetID(char const * prefix, int page_number, int counter) { | |
| static std::string GetID(const char *prefix, int page_number, int counter) { |
There was a problem hiding this comment.
The original code is technically fine, but my suggested change uses the coding style which is typical for Tesseract code.
When running tesseract on a list of images and validating the alto output using the schema (https://www.loc.gov/alto/v3/alto-3-0.xsd), I had the following error:
Digging into the type of
BlockTypeID, I found axsd:IDrestriction - meaning theBlockTypeIDcan only exist once in the whole document / is unique per document.I incorporated the current
page_numberinto the IDs thus resolving the issue.