Fix SST parsing for strings spanning Continue record boundaries - #2
Open
mattf-intex wants to merge 1 commit into
Open
Fix SST parsing for strings spanning Continue record boundaries#2mattf-intex wants to merge 1 commit into
mattf-intex wants to merge 1 commit into
Conversation
The SST decoder was incorrectly concatenating Continue record bytes without accounting for the grbit byte that prefixes string continuations. Per MS-XLS spec (2.4.265 SST, 2.4.39 Continue), when a string spans a Continue record boundary, the Continue record begins with a 1-byte grbit field indicating the encoding of the remaining characters. This caused "sst get" index errors when reading XLS files with large shared string tables that span multiple Continue records. The fix rewrites the decode method to: - Track position across SST and Continue record boundaries - Skip the grbit byte at Continue boundaries when parsing strings - Handle encoding changes (compressed vs UTF-16) at boundaries Or a shorter version: Fix SST parsing to handle Continue record boundaries correctly Strings spanning Continue records were corrupted because the grbit byte at Continue boundaries was being included as string data. Now properly skips grbit and handles encoding changes per MS-XLS spec. Fixes "sst get" errors on XLS files with large shared string tables.
ketbra
pushed a commit
to ketbra/xlrd
that referenced
this pull request
Jan 21, 2026
Incorporates fix from PR #2 (PleaseDont#2) Problem: The SST (Shared String Table) decoder was incorrectly handling string data that spans multiple Continue records. When strings span across Continue record boundaries, each continuation starts with a 1-byte grbit field indicating character encoding. The old implementation concatenated all Continue record bytes without accounting for these grbit bytes, treating them as string data instead of metadata. This caused: - String corruption in large shared string tables - "sst get" index errors when processing XLS files - Incorrect string decoding when encoding changed at record boundaries Solution: Complete rewrite of SST parsing with new SstReader struct: - Tracks position across main SST record and Continue records - Properly skips grbit byte when transitioning to Continue records mid-string - Handles encoding transitions (compressed ↔ UTF-16) at record boundaries - Manually iterates through strings with proper boundary awareness Key changes (src/record/sst.rs): - Add SstReader struct with methods: read_byte(), read_u16(), read_i32() - Add read_string_bytes() to handle Continue boundaries correctly - Rewrite decode() to use SstReader instead of naive concatenation - Remove XLUnicodeRichExtendedString struct (no longer needed) - Remove until_eof import (now doing manual iteration) Test Results: **71/71 files now pass (100%)** - Previously: 70/71 (98.6%) - 55982.xls failed with "sst get" error - Now: 71/71 (100%) - ALL files parse successfully! This fix is critical for correctly reading XLS files with large shared string tables that require multiple BIFF records.
mattf-intex
pushed a commit
to mattf-intex/xlrd
that referenced
this pull request
Feb 12, 2026
Ported python xlrd tests to rust
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The SST decoder was incorrectly concatenating Continue record bytes without accounting for the grbit byte that prefixes string continuations. Per MS-XLS spec (2.4.265 SST, 2.4.39 Continue), when a string spans a Continue record boundary, the Continue record begins with a 1-byte grbit field indicating the encoding of the remaining characters.
This caused "sst get" index errors when reading XLS files with large shared string tables that span multiple Continue records.
The fix rewrites the decode method to:
Or a shorter version:
Fix SST parsing to handle Continue record boundaries correctly
Strings spanning Continue records were corrupted because the grbit byte at Continue boundaries was being included as string data. Now properly skips grbit and handles encoding changes per MS-XLS spec.
Fixes "sst get" errors on XLS files with large shared string tables.