staging: MHTML support (mirror of upstream #149) #7
There are no files selected for viewing
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
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
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
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,10 @@ | ||
| import test from 'node:test' | ||
| import assert from 'node:assert/strict' | ||
| import { formatFromBytes, formatFromExtension, toMarkdownBytes } from './index.js' | ||
|
|
||
| test('standalone HTML is exposed through the Node binding', async () => { | ||
| const input = Buffer.from('<!doctype html><h1>Hello</h1><p><b>world</b></p>') | ||
| assert.equal(formatFromExtension('html'), 'html') | ||
| assert.equal(formatFromBytes(input), 'html') | ||
| assert.equal(await toMarkdownBytes(input), '# Hello\n\n**world**\n') | ||
| }) |
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
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,18 @@ | ||
| import test from 'node:test' | ||
| import assert from 'node:assert/strict' | ||
| import { formatFromBytes, formatFromExtension, toMarkdownBytes } from './index.js' | ||
|
|
||
| const input = Buffer.from( | ||
| 'Snapshot-Content-Location: https://example.test/page\r\n' + | ||
| 'MIME-Version: 1.0\r\n' + | ||
| 'Content-Type: multipart/related; type="text/html"; boundary="b"\r\n\r\n' + | ||
| '--b\r\nContent-Type: text/html\r\n\r\n' + | ||
| '<!doctype html><h1>Hello MHTML</h1>\r\n--b--\r\n' | ||
| ) | ||
|
|
||
| test('MHTML is exposed through the Node binding', async () => { | ||
| assert.equal(formatFromExtension('mhtml'), 'mhtml') | ||
| assert.equal(formatFromExtension('mht'), 'mhtml') | ||
| assert.equal(formatFromBytes(input), 'mhtml') | ||
| assert.equal(await toMarkdownBytes(input), '# Hello MHTML\n') | ||
| }) |
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
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
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -67,7 +67,7 @@ create_exception!( | |
| /// Format names, as the extension that identifies each format. Container | ||
| /// variants that share a parser (`.docm`, `.xlsm`, `.ppsx`, ...) map onto | ||
| /// these via `format_from_bytes` or `format_from_extension`. | ||
| const FORMATS: [(&str, anydoc::Format); 14] = [ | ||
| ("doc", anydoc::Format::Doc), | ||
| ("docx", anydoc::Format::Docx), | ||
| ("odt", anydoc::Format::Odt), | ||
|
|
@@ -76,6 +76,8 @@ const FORMATS: [(&str, anydoc::Format); 12] = [ | |
| ("pptx", anydoc::Format::Pptx), | ||
| ("rtf", anydoc::Format::Rtf), | ||
| ("epub", anydoc::Format::Epub), | ||
| ("html", anydoc::Format::Html), | ||
| ("mhtml", anydoc::Format::Mhtml), | ||
| ("xlsx", anydoc::Format::Excel), | ||
| ("ods", anydoc::Format::Ods), | ||
| ("odp", anydoc::Format::Odp), | ||
|
|
@@ -139,9 +141,10 @@ fn convert_error(py: Python<'_>, error: anydoc::ConvertError) -> PyErr { | |
| } | ||
|
|
||
| /// Detect the format from the content itself: the signature and identity each | ||
| /// container specification designates (PDF header, RTF open group, MIME HTML | ||
| /// aggregate, OLE stream names, ZIP package mimetype/content types). | ||
| /// Plain-text formats (CSV) carry no signature and return `None`; so does | ||
| /// anything unrecognized. | ||
| #[pyfunction] | ||
| fn format_from_bytes(data: Vec<u8>) -> Option<&'static str> { | ||
| anydoc::Format::from_bytes(&data).map(format_name) | ||
|
|
||
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,15 @@ | ||
| import unittest | ||
|
|
||
| import anydoc | ||
|
|
||
|
|
||
| class HtmlBindingTests(unittest.TestCase): | ||
| def test_standalone_html_is_exposed(self): | ||
| data = b'<!doctype html><h1>Hello</h1><p><b>world</b></p>' | ||
| self.assertEqual(anydoc.format_from_extension('html'), 'html') | ||
| self.assertEqual(anydoc.format_from_bytes(data), 'html') | ||
| self.assertEqual(anydoc.to_markdown_bytes(data), '# Hello\n\n**world**\n') | ||
|
|
||
|
|
||
| if __name__ == '__main__': | ||
| unittest.main() |
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,22 @@ | ||
| import unittest | ||
| import anydoc | ||
|
|
||
| INPUT = ( | ||
| b'Snapshot-Content-Location: https://example.test/page\r\n' | ||
| b'MIME-Version: 1.0\r\n' | ||
| b'Content-Type: multipart/related; type="text/html"; boundary="b"\r\n\r\n' | ||
| b'--b\r\nContent-Type: text/html\r\n\r\n' | ||
| b'<!doctype html><h1>Hello MHTML</h1>\r\n--b--\r\n' | ||
| ) | ||
|
|
||
|
|
||
| class MhtmlTests(unittest.TestCase): | ||
| def test_mhtml_binding(self): | ||
| self.assertEqual(anydoc.format_from_extension('mhtml'), 'mhtml') | ||
| self.assertEqual(anydoc.format_from_extension('mht'), 'mhtml') | ||
| self.assertEqual(anydoc.format_from_bytes(INPUT), 'mhtml') | ||
| self.assertEqual(anydoc.to_markdown_bytes(INPUT), '# Hello MHTML\n') | ||
|
There was a problem hiding this comment. P2: This binding test only exercises the happy path (detection + conversion) and none of the MHTML-specific behavior this PR adds: the base64-reserve / quoted-printable bounds / nesting-depth / media-type-guard resource limits, Content-ID/Content-Location/base-href resolution, or the MHTML-over-HTML detection preference. A regression in error propagation through the binding (e.g. Prompt for AI agents
|
||
|
|
||
|
|
||
| if __name__ == '__main__': | ||
| unittest.main() | ||
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,7 +1,9 @@ | ||
| //! Content-based format detection. | ||
| //! | ||
| //! Identifies the format primarily from what each specification designates as | ||
| //! the container's identity. Standalone HTML is the one intentional content | ||
| //! heuristic: after an optional BOM/leading whitespace, a leading HTML5 | ||
| //! `<!DOCTYPE ... html>` or `<html>` marker identifies `Format::Html`. | ||
| //! | ||
| //! - PDF: the `%PDF-` header (ISO 32000; implementations accept leading | ||
| //! junk, bounded here at 1024 bytes). | ||
|
|
@@ -41,12 +43,117 @@ pub(crate) fn from_bytes(bytes: &[u8]) -> Option<Format> { | |
| if bytes.starts_with(b"PK\x03\x04") { | ||
| return detect_zip(bytes); | ||
| } | ||
| if looks_like_html(bytes) { | ||
| return Some(Format::Html); | ||
| } | ||
| if bytes[..bytes.len().min(1024)].windows(5).any(|w| w == b"%PDF-") { | ||
| return Some(Format::Pdf); | ||
| } | ||
| None | ||
| } | ||
|
|
||
| fn looks_like_html(bytes: &[u8]) -> bool { | ||
| if let Some(rest) = bytes.strip_prefix(&[0xFF, 0xFE]) { | ||
| return looks_like_utf16_html(rest, true); | ||
| } | ||
| if let Some(rest) = bytes.strip_prefix(&[0xFE, 0xFF]) { | ||
| return looks_like_utf16_html(rest, false); | ||
| } | ||
| let bytes = bytes.strip_prefix(&[0xEF, 0xBB, 0xBF]).unwrap_or(bytes); | ||
| looks_like_ascii_html(bytes) | ||
| } | ||
|
|
||
| fn looks_like_utf16_html(bytes: &[u8], little_endian: bool) -> bool { | ||
| let (pairs, _) = bytes.as_chunks::<2>(); | ||
| let mut index = 0; | ||
| while pairs.get(index).is_some_and(|pair| { | ||
| utf16_ascii_unit(*pair, little_endian).is_some_and(|b| b.is_ascii_whitespace()) | ||
| }) { | ||
| index += 1; | ||
| } | ||
|
|
||
| if utf16_html_prefix(pairs, index, little_endian, b"<html") { | ||
| return true; | ||
| } | ||
|
|
||
| const DOCTYPE: &[u8] = b"<!doctype"; | ||
| if !utf16_prefix_eq_ignore_ascii_case(pairs, index, little_endian, DOCTYPE) { | ||
| return false; | ||
| } | ||
| index += DOCTYPE.len(); | ||
| if !pairs.get(index).is_some_and(|pair| { | ||
| utf16_ascii_unit(*pair, little_endian).is_some_and(|b| b.is_ascii_whitespace()) | ||
| }) { | ||
| return false; | ||
| } | ||
| while pairs.get(index).is_some_and(|pair| { | ||
| utf16_ascii_unit(*pair, little_endian).is_some_and(|b| b.is_ascii_whitespace()) | ||
| }) { | ||
| index += 1; | ||
| } | ||
| utf16_html_prefix(pairs, index, little_endian, b"html") | ||
| } | ||
|
|
||
| fn utf16_ascii_unit(pair: [u8; 2], little_endian: bool) -> Option<u8> { | ||
| let unit = if little_endian { u16::from_le_bytes(pair) } else { u16::from_be_bytes(pair) }; | ||
| (unit <= 0x7F).then_some(unit as u8) | ||
| } | ||
|
|
||
| fn utf16_prefix_eq_ignore_ascii_case( | ||
| pairs: &[[u8; 2]], | ||
| start: usize, | ||
| little_endian: bool, | ||
| prefix: &[u8], | ||
| ) -> bool { | ||
| let Some(slice) = pairs.get(start..start + prefix.len()) else { | ||
| return false; | ||
| }; | ||
| slice.iter().zip(prefix).all(|(pair, expected)| { | ||
| utf16_ascii_unit(*pair, little_endian) | ||
| .is_some_and(|byte| byte.eq_ignore_ascii_case(expected)) | ||
| }) | ||
| } | ||
|
|
||
| fn utf16_html_prefix(pairs: &[[u8; 2]], start: usize, little_endian: bool, prefix: &[u8]) -> bool { | ||
| utf16_prefix_eq_ignore_ascii_case(pairs, start, little_endian, prefix) | ||
| && pairs.get(start + prefix.len()).is_none_or(|pair| { | ||
| utf16_ascii_unit(*pair, little_endian) | ||
| .is_some_and(|b| b.is_ascii_whitespace() || matches!(b, b'>' | b'/')) | ||
| }) | ||
| } | ||
|
|
||
| fn looks_like_ascii_html(bytes: &[u8]) -> bool { | ||
| let bytes = bytes.trim_ascii_start(); | ||
| html_doctype_prefix(bytes) || html_prefix(bytes, b"<html") | ||
| } | ||
|
|
||
| fn html_doctype_prefix(bytes: &[u8]) -> bool { | ||
| const DOCTYPE: &[u8] = b"<!doctype"; | ||
| let Some(head) = bytes.get(..DOCTYPE.len()) else { | ||
| return false; | ||
| }; | ||
| if !head.eq_ignore_ascii_case(DOCTYPE) { | ||
| return false; | ||
| } | ||
| let Some(rest) = bytes.get(DOCTYPE.len()..) else { | ||
| return false; | ||
| }; | ||
| if !rest.first().is_some_and(u8::is_ascii_whitespace) { | ||
| return false; | ||
| } | ||
| html_prefix(rest.trim_ascii_start(), b"html") | ||
| } | ||
|
|
||
| fn html_prefix(bytes: &[u8], prefix: &[u8]) -> bool { | ||
| let Some(head) = bytes.get(..prefix.len()) else { | ||
| return false; | ||
| }; | ||
| head.eq_ignore_ascii_case(prefix) | ||
| && bytes | ||
| .get(prefix.len()) | ||
| .is_none_or(|b| b.is_ascii_whitespace() || matches!(b, b'>' | b'/')) | ||
| } | ||
|
|
||
| /// Classify an OLE compound file by its mandated content stream. Encrypted | ||
| /// OOXML packages (`EncryptedPackage`) stay `None`: the inner format is | ||
| /// unknowable, and the frontend reports `Encrypted` precisely. | ||
|
|
@@ -236,6 +343,8 @@ mod tests { | |
| junk.extend_from_slice(b"%PDF-1.4"); | ||
| assert_eq!(from_bytes(&junk), Some(Format::Pdf)); | ||
| assert_eq!(from_bytes(b"{\\rtf1\\ansi hi}"), Some(Format::Rtf)); | ||
| assert_eq!(from_bytes(b"<!DOCTYPE html><html></html>"), Some(Format::Html)); | ||
| assert_eq!(from_bytes(b"\xEF\xBB\xBF <HTML><body>x</body></HTML>"), Some(Format::Html)); | ||
| assert_eq!(from_bytes(b"a,b,c\n1,2,3\n"), None); | ||
| assert_eq!(from_bytes(b""), None); | ||
| } | ||
|
|
||
Oops, something went wrong.
Oops, something went wrong.
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.
Uh oh!
There was an error while loading. Please reload this page.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
P2: Type checkers still reject
"html"and"mhtml"when callers use the publicanydoc.to_markdown_bytesoranydoc.to_documentAPIs. Add both literals topython/anydoc/__init__.py's publicFormatalias as well.Prompt for AI agents