World's first open-source enterprise world model.
Rust agent-memorybitemporalgraphragknowledge-baseknowledge-graphllmontologypgvectorpostgresqlragrustself-hostedsemantic-searchtemporal-knowledge-graphworld-model
A slide table keeps its rows and columns (#1023) (#1042)
Coderfu62e49ed72f6a
diff --git a/crates/utopia-ingest/src/parsers.rs b/crates/utopia-ingest/src/parsers.rsindex c60e7fa9..1194aac6 100644--- a/crates/utopia-ingest/src/parsers.rs+++ b/crates/utopia-ingest/src/parsers.rs@@ -480,7 +480,8 @@ pub(crate) fn docx_xml_to_text(Ok(out)}-/// PPTX: extract a:t text in the presentation's logical slide order.+/// PPTX: extract a:t text in the presentation's logical slide order. Tables under a:tbl+/// become Markdown grids through the same renderer as DOCX and spreadsheet tables.pub fn pptx(bytes: &[u8]) -> anyhow::Result<String> {let mut archive =zip::ZipArchive::new(Cursor::new(bytes.to_vec())).context("Failed to unzip pptx")?;@@ -510,7 +511,7 @@ pub fn pptx(bytes: &[u8]) -> anyhow::Result<String> {let mut out = String::new();for (num, name) in slides {let xml = pptx_part(&mut archive, &name)?;- let text = extract_xml_text(&xml, "a:t", "a:p", "a:br")?;+ let text = pptx_xml_to_text(&xml)?;if !text.trim().is_empty() {out.push_str(&format!("\n## Slide {num}\n{text}\n"));}@@ -1015,45 +1016,159 @@ fn read_zip_entry(Ok(content)}-/// 从 OOXML 里抽取 `text_tag`(如 a:t)内的文本,遇 `para_tag`(如 a:p)结束换行,-/// 遇 `break_tag`(如 a:br)也换行。-fn extract_xml_text(- xml: &str,- text_tag: &str,- para_tag: &str,- break_tag: &str,-) -> anyhow::Result<String> {+/// 从 PPTX 的 slide XML 里读文字。表外的 `a:p`/`a:br` 仍按原来的换行规则;+/// `a:tbl` 里的每个 `a:tr` 收成一行,每个 `a:tc` 收成一格,格内段落用空格连接。+fn pptx_xml_to_text(xml: &str) -> anyhow::Result<String> {+ #[derive(Default)]+ struct TableCell {+ text: String,+ span: usize,+ horizontal_merge: bool,+ vertical_merge: bool,+ }++ fn attr(e: &quick_xml::events::BytesStart<'_>, name: &str) -> Option<String> {+ e.attributes()+ .flatten()+ .find(|a| a.key.as_ref() == name)+ .map(|a| a.value.to_string())+ }++ fn truthy(e: &quick_xml::events::BytesStart<'_>, name: &str) -> bool {+ matches!(attr(e, name).as_deref(), Some("1" | "true" | "on"))+ }+let mut reader = Reader::from_str(xml);let mut out = String::new();let mut in_text = false;+ let mut table_depth = 0usize;+ let mut rows: Vec<GridRow> = Vec::new();+ let mut row: GridRow = Vec::new();+ let mut cell: Option<TableCell> = None;+ let mut first_is_header = false;loop {match reader.read_event() {- Ok(Event::Start(e)) if e.name().as_ref() == text_tag => in_text = true,- // 段落里换的行(Shift+Enter)不是新段落,是两个 run 之间的一个 `<a:br>`。丢了它,- // 标题「Q1」换行「2024」读成「Q12024」,季度和年份一起没了(Word 格子里的同一件事- // 见 #813)- Ok(Event::Start(e) | Event::Empty(e)) if e.name().as_ref() == break_tag => {- out.push('\n');- }- Ok(Event::End(e)) => {- let name = e.name();- if name.as_ref() == text_tag {- in_text = false;- } else if name.as_ref() == para_tag {- out.push('\n');+ Ok(Event::Start(e)) => match e.name().as_ref() {+ "a:tbl" => {+ table_depth += 1;+ if table_depth == 1 {+ rows.clear();+ row.clear();+ cell = None;+ first_is_header = false;+ }}- }+ "a:tblPr" if table_depth == 1 => {+ first_is_header = truthy(&e, "firstRow");+ }+ "a:tr" if table_depth == 1 => row.clear(),+ "a:tc" if table_depth == 1 => {+ cell = Some(TableCell {+ span: 1,+ ..TableCell::default()+ });+ }+ "a:tcPr" if table_depth == 1 => {+ if let Some(c) = cell.as_mut() {+ if let Some(span) = attr(&e, "gridSpan").and_then(|v| v.parse().ok()) {+ c.span = span;+ }+ c.horizontal_merge = truthy(&e, "hMerge");+ c.vertical_merge = truthy(&e, "vMerge");+ }+ }+ "a:t" => {+ if cell.is_some() || table_depth == 0 {+ in_text = true;+ }+ }+ "a:br" => match cell.as_mut() {+ Some(c) => c.text.push(' '),+ None if table_depth == 0 => out.push('\n'),+ _ => {}+ },+ _ => {}+ },+ Ok(Event::Empty(e)) => match e.name().as_ref() {+ "a:tblPr" if table_depth == 1 => {+ first_is_header = truthy(&e, "firstRow");+ }+ "a:tcPr" if table_depth == 1 => {+ if let Some(c) = cell.as_mut() {+ if let Some(span) = attr(&e, "gridSpan").and_then(|v| v.parse().ok()) {+ c.span = span;+ }+ c.horizontal_merge = truthy(&e, "hMerge");+ c.vertical_merge = truthy(&e, "vMerge");+ }+ }+ "a:br" => match cell.as_mut() {+ Some(c) => c.text.push(' '),+ None if table_depth == 0 => out.push('\n'),+ _ => {}+ },+ "a:tab" => {+ if let Some(c) = cell.as_mut() {+ c.text.push(' ');+ }+ }+ _ => {}+ },+ Ok(Event::End(e)) => match e.name().as_ref() {+ "a:t" => in_text = false,+ "a:p" => match cell.as_mut() {+ Some(c) => c.text.push(' '),+ None if table_depth == 0 => out.push('\n'),+ _ => {}+ },+ "a:tc" if table_depth == 1 => {+ if let Some(mut c) = cell.take() {+ if !c.horizontal_merge {+ if c.vertical_merge {+ c.text.clear();+ }+ row.push((c.text, c.span.max(1), 0));+ }+ }+ }+ "a:tr" if table_depth == 1 => rows.push(std::mem::take(&mut row)),+ "a:tbl" => {+ table_depth = table_depth.saturating_sub(1);+ if table_depth == 0 {+ if let Some(md) = crate::table::render_grid(&rows, first_is_header) {+ out.push('\n');+ out.push_str(&md);+ out.push_str("\n\n");+ }+ rows.clear();+ row.clear();+ cell = None;+ first_is_header = false;+ }+ }+ _ => {}+ },Ok(Event::Text(t)) if in_text => {- out.push_str(&t.xml_content(quick_xml::XmlVersion::Implicit1_0));+ let text = t.xml_content(quick_xml::XmlVersion::Implicit1_0);+ match cell.as_mut() {+ Some(c) => c.text.push_str(&text),+ None => out.push_str(&text),+ }}Ok(Event::CData(t)) if in_text => {- out.push_str(&t.xml_content(quick_xml::XmlVersion::Implicit1_0));+ let text = t.xml_content(quick_xml::XmlVersion::Implicit1_0);+ match cell.as_mut() {+ Some(c) => c.text.push_str(&text),+ None => out.push_str(&text),+ }}Ok(Event::GeneralRef(e)) if in_text => {let reference = format!("&{};", e.into_inner());- match quick_xml::escape::unescape(&reference) {- Ok(s) => out.push_str(&s),- Err(_) => out.push_str(&reference),+ let text = quick_xml::escape::unescape(&reference)+ .unwrap_or(std::borrow::Cow::Borrowed(&reference));+ match cell.as_mut() {+ Some(c) => c.text.push_str(&text),+ None => out.push_str(&text),}}Ok(Event::Eof) => break,diff --git a/crates/utopia-ingest/tests/pptx_tables.rs b/crates/utopia-ingest/tests/pptx_tables.rsnew file mode 100644index 00000000..2f7eadbb--- /dev/null+++ b/crates/utopia-ingest/tests/pptx_tables.rs@@ -0,0 +1,217 @@+use std::io::{Cursor, Write};++fn package(parts: &[(String, String)]) -> Vec<u8> {+ let mut zip = zip::ZipWriter::new(Cursor::new(Vec::new()));+ for (name, xml) in parts {+ zip.start_file(name, zip::write::SimpleFileOptions::default())+ .unwrap();+ zip.write_all(xml.as_bytes()).unwrap();+ }+ zip.finish().unwrap().into_inner()+}++fn deck(body: &str) -> Vec<u8> {+ package(&[+ ("[Content_Types].xml".into(), r#"<Types xmlns="http://schemas.openxmlformats.org/package/2006/content-types"><Default Extension="rels" ContentType="application/vnd.openxmlformats-package.relationships+xml"/><Override PartName="/ppt/presentation.xml" ContentType="application/vnd.openxmlformats-officedocument.presentationml.presentation.main+xml"/><Override PartName="/ppt/slides/slide1.xml" ContentType="application/vnd.openxmlformats-officedocument.presentationml.slide+xml"/></Types>"#.into()),+ ("_rels/.rels".into(), r#"<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument" Target="ppt/presentation.xml"/></Relationships>"#.into()),+ ("ppt/presentation.xml".into(), r#"<p:presentation xmlns:p="http://schemas.openxmlformats.org/presentationml/2006/main" xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships"><p:sldIdLst><p:sldId id="256" r:id="rId2"/></p:sldIdLst></p:presentation>"#.into()),+ ("ppt/_rels/presentation.xml.rels".into(), r#"<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships"><Relationship Id="rId2" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/slide" Target="slides/slide1.xml"/></Relationships>"#.into()),+ ("ppt/slides/slide1.xml".into(), format!(r#"<p:sld xmlns:p="http://schemas.openxmlformats.org/presentationml/2006/main" xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main"><p:cSld><p:spTree>{body}</p:spTree></p:cSld></p:sld>"#)),+ ])+}++fn parse_slide(body: &str) -> String {+ utopia_ingest::parse("deck.pptx", &deck(body)).unwrap().text+}++fn text_box(text: &str) -> String {+ format!(+ r#"<p:sp><p:txBody><a:bodyPr/><a:lstStyle/><a:p><a:r><a:t>{text}</a:t></a:r></a:p></p:txBody></p:sp>"#+ )+}++fn cell(text: &str) -> String {+ cell_with_props(text, "")+}++fn cell_with_props(text: &str, props: &str) -> String {+ format!(+ r#"<a:tc><a:txBody><a:bodyPr/><a:lstStyle/><a:p><a:r><a:t>{text}</a:t></a:r></a:p></a:txBody><a:tcPr {props}/></a:tc>"#+ )+}++fn cell_with_paragraphs(paragraphs: &[&str]) -> String {+ let paragraphs = paragraphs+ .iter()+ .map(|text| format!(r#"<a:p><a:r><a:t>{text}</a:t></a:r></a:p>"#))+ .collect::<String>();+ format!(r#"<a:tc><a:txBody><a:bodyPr/><a:lstStyle/>{paragraphs}</a:txBody><a:tcPr/></a:tc>"#)+}++fn empty_cell(props: &str) -> String {+ format!(r#"<a:tc><a:txBody><a:bodyPr/><a:lstStyle/><a:p/></a:txBody><a:tcPr {props}/></a:tc>"#)+}++fn row(cells: &[String]) -> String {+ format!("<a:tr>{}</a:tr>", cells.concat())+}++fn table(first_row: bool, rows: &[String]) -> String {+ let first_row = if first_row { r#" firstRow="1""# } else { "" };+ format!(+ r#"<p:graphicFrame><a:graphic><a:graphicData uri="http://schemas.openxmlformats.org/drawingml/2006/table"><a:tbl><a:tblPr{first_row}/><a:tblGrid/><rows/></a:tbl></a:graphicData></a:graphic></p:graphicFrame>"#+ )+ .replace("<rows/>", &rows.concat())+}++#[test]+fn a_three_by_four_slide_table_keeps_its_header_and_rows() {+ let text = parse_slide(&table(+ true,+ &[+ row(&[cell("Quarter"), cell("Revenue"), cell("YoY")]),+ row(&[cell("Q1"), cell("1,200"), cell("+12%")]),+ row(&[cell("Q2"), cell("1,350"), cell("+9%")]),+ row(&[cell("Q3"), cell("1,410"), cell("+7%")]),+ ],+ ));++ assert!(+ text.contains(+ "| Quarter | Revenue | YoY |\n| --- | --- | --- |\n| Q1 | 1,200 | +12% |\n| Q2 | 1,350 | +9% |\n| Q3 | 1,410 | +7% |"+ ),+ "{text}"+ );+}++#[test]+fn a_title_and_text_around_a_slide_table_keep_their_lines() {+ let text = parse_slide(&format!(+ "{}{}{}",+ text_box("2024 年各季度营收"),+ table(+ true,+ &[+ row(&[cell("Quarter"), cell("Revenue")]),+ row(&[cell("Q1"), cell("1,200")]),+ ],+ ),+ text_box("来源:财务部"),+ ));++ let title = text.find("2024 年各季度营收").expect(&text);+ let table_start = text.find("| Quarter | Revenue |").expect(&text);+ let source = text.find("来源:财务部").expect(&text);+ assert!(title < table_start && table_start < source, "{text}");+}++#[test]+fn grid_span_and_horizontal_merge_keep_the_grid_shape() {+ let text = parse_slide(&table(+ true,+ &[+ row(&[+ cell_with_props("Summary", r#"gridSpan="2""#),+ cell_with_props("COVERED", r#"hMerge="1""#),+ cell("Total"),+ ]),+ row(&[cell("Q1"), cell("1,200"), cell("1,350")]),+ ],+ ));++ assert!(+ text.contains("| | Summary | Total |\n| --- | --- | --- |\n| Q1 | 1,200 | 1,350 |"),+ "{text}"+ );+ assert!(!text.contains("COVERED"), "{text}");+}++#[test]+fn a_vertical_merge_continuation_stays_an_empty_cell() {+ let text = parse_slide(&table(+ true,+ &[+ row(&[cell("Quarter"), cell("Revenue"), cell("YoY")]),+ row(&[cell("Q1"), cell("1,200"), cell("+12%")]),+ row(&[cell("Q2"), cell("1,350"), empty_cell(r#"vMerge="1""#)]),+ row(&[cell("Q3"), cell("1,410"), cell("+7%")]),+ ],+ ));++ assert!(+ text.contains(+ "| Quarter | Revenue | YoY |\n| --- | --- | --- |\n| Q1 | 1,200 | +12% |\n| Q2 | 1,350 | |\n| Q3 | 1,410 | +7% |"+ ),+ "{text}"+ );+}++#[test]+fn first_row_marks_the_first_row_as_a_header() {+ let text = parse_slide(&table(+ true,+ &[+ row(&[cell("2024"), cell("2025")]),+ row(&[cell("Revenue"), cell("1,200")]),+ ],+ ));++ assert!(+ text.contains("| 2024 | 2025 |\n| --- | --- |\n| Revenue | 1,200 |"),+ "{text}"+ );+}++#[test]+fn without_first_row_rows_are_classified_by_their_content() {+ let text = parse_slide(&table(+ false,+ &[+ row(&[cell("2024"), cell("2025")]),+ row(&[cell("Revenue"), cell("1,200")]),+ ],+ ));++ assert!(+ text.contains("| | |\n| --- | --- |\n| 2024 | 2025 |\n| Revenue | 1,200 |"),+ "{text}"+ );+}++#[test]+fn a_slide_can_hold_two_tables_without_joining_them() {+ let text = parse_slide(&format!(+ "{}{}{}",+ table(+ true,+ &[row(&[cell("A"), cell("B")]), row(&[cell("1"), cell("2")]),],+ ),+ text_box("Between"),+ table(+ true,+ &[row(&[cell("C"), cell("D")]), row(&[cell("3"), cell("4")]),],+ ),+ ));++ assert_eq!(text.matches("| --- | --- |").count(), 2, "{text}");+ let first = text.find("| A | B |").expect(&text);+ let between = text.find("Between").expect(&text);+ let second = text.find("| C | D |").expect(&text);+ assert!(first < between && between < second, "{text}");+}++#[test]+fn paragraphs_in_a_cell_are_joined_by_spaces() {+ let text = parse_slide(&table(+ true,+ &[+ row(&[cell("Region"), cell("Sales")]),+ row(&[cell_with_paragraphs(&["North", "America"]), cell("1,200")]),+ ],+ ));++ assert!(+ text.contains("| Region | Sales |\n| --- | --- |\n| North America | 1,200 |"),+ "{text}"+ );+}