diff options
| -rw-r--r-- | mammoth/conversion.py | 104 | ||||
| -rw-r--r-- | mammoth/documents.py | 3 | ||||
| -rw-r--r-- | mammoth/lists.py | 6 | ||||
| -rw-r--r-- | setup.py | 2 | ||||
| -rw-r--r-- | tests/conversion_tests.py | 29 |
5 files changed, 109 insertions, 35 deletions
diff --git a/mammoth/conversion.py b/mammoth/conversion.py index 489c74a..cbe207a 100644 --- a/mammoth/conversion.py +++ b/mammoth/conversion.py @@ -4,8 +4,11 @@ from __future__ import unicode_literals from functools import partial -from . import documents, results, html_paths, images, writers, html, lists +import cobble + +from . import documents, results, html_paths, images, writers, html from .docx.files import InvalidFileReferenceError +from .lists import find_index def convert_document_element_to_html(element, @@ -42,14 +45,23 @@ def convert_document_element_to_html(element, note_references=[], comments=comments, ) - nodes = converter.visit(element) + context = _ConversionContext(is_table_header=False) + nodes = converter.visit(element, context) writer = writers.writer(output_format) html.write(writer, html.collapse(html.strip_empty(nodes))) return results.Result(writer.as_string(), messages) -class _DocumentConverter(documents.ElementVisitor): +@cobble.data +class _ConversionContext(object): + is_table_header = cobble.field() + + def copy(self, **kwargs): + return cobble.copy(self, **kwargs) + + +class _DocumentConverter(documents.element_visitor(args=1)): def __init__(self, messages, style_map, convert_image, id_prefix, ignore_empty_paragraphs, note_references, comments): self._messages = messages self._style_map = style_map @@ -60,27 +72,31 @@ class _DocumentConverter(documents.ElementVisitor): self._convert_image = convert_image self._comments = comments - def visit_image(self, image): + def visit_image(self, image, context): try: return self._convert_image(image) except InvalidFileReferenceError as error: self._messages.append(results.warning(str(error))) return [] - def visit_document(self, document): - nodes = self._visit_all(document.children) + def visit_document(self, document, context): + nodes = self._visit_all(document.children, context) notes = [ document.notes.resolve(reference) for reference in self._note_references ] - notes_list = html.element("ol", {}, self._visit_all(notes)) - comments = html.element("dl", {}, lists.flat_map(self.visit_comment, self._referenced_comments)) + notes_list = html.element("ol", {}, self._visit_all(notes, context)) + comments = html.element("dl", {}, [ + html_node + for referenced_comment in self._referenced_comments + for html_node in self.visit_comment(referenced_comment, context) + ]) return nodes + [notes_list, comments] - def visit_paragraph(self, paragraph): + def visit_paragraph(self, paragraph, context): def children(): - content = self._visit_all(paragraph.children) + content = self._visit_all(paragraph.children, context) if self._ignore_empty_paragraphs: return content else: @@ -90,8 +106,8 @@ class _DocumentConverter(documents.ElementVisitor): return html_path.wrap(children) - def visit_run(self, run): - nodes = lambda: self._visit_all(run.children) + def visit_run(self, run, context): + nodes = lambda: self._visit_all(run.children, context) paths = [] if run.is_strikethrough: paths.append(self._find_style_for_run_property("strikethrough", default="s")) @@ -123,21 +139,21 @@ class _DocumentConverter(documents.ElementVisitor): return html_paths.empty - def visit_text(self, text): + def visit_text(self, text, context): return [html.text(text.value)] - def visit_hyperlink(self, hyperlink): + def visit_hyperlink(self, hyperlink, context): if hyperlink.anchor is None: href = hyperlink.href else: href = "#{0}".format(self._html_id(hyperlink.anchor)) - nodes = self._visit_all(hyperlink.children) + nodes = self._visit_all(hyperlink.children, context) return [html.collapsible_element("a", {"href": href}, nodes)] - def visit_bookmark(self, bookmark): + def visit_bookmark(self, bookmark, context): element = html.collapsible_element( "a", {"id": self._html_id(bookmark.name)}, @@ -145,31 +161,49 @@ class _DocumentConverter(documents.ElementVisitor): return [element] - def visit_tab(self, tab): + def visit_tab(self, tab, context): return [html.text("\t")] - def visit_table(self, table): - return [html.element("table", {}, self._visit_all(table.children))] + def visit_table(self, table, context): + body_index = find_index(lambda child: not child.is_header, table.children) + if body_index is None: + body_index = len(table.children) + + if body_index == 0: + children = self._visit_all(table.children, context.copy(is_table_header=False)) + else: + head_rows = self._visit_all(table.children[:body_index], context.copy(is_table_header=True)) + body_rows = self._visit_all(table.children[body_index:], context.copy(is_table_header=False)) + children = [ + html.element("thead", {}, head_rows), + html.element("tbody", {}, body_rows), + ] + + return [html.element("table", {}, children)] - def visit_table_row(self, table_row): - return [html.element("tr", {}, self._visit_all(table_row.children))] + def visit_table_row(self, table_row, context): + return [html.element("tr", {}, self._visit_all(table_row.children, context))] - def visit_table_cell(self, table_cell): + def visit_table_cell(self, table_cell, context): + if context.is_table_header: + tag_name = "th" + else: + tag_name = "td" attributes = {} if table_cell.colspan != 1: attributes["colspan"] = str(table_cell.colspan) if table_cell.rowspan != 1: attributes["rowspan"] = str(table_cell.rowspan) - nodes = [html.force_write] + self._visit_all(table_cell.children) + nodes = [html.force_write] + self._visit_all(table_cell.children, context) return [ - html.element("td", attributes, nodes) + html.element(tag_name, attributes, nodes) ] - def visit_break(self, break_): + def visit_break(self, break_, context): return self._find_html_path_for_break(break_).wrap(lambda: []) @@ -183,7 +217,7 @@ class _DocumentConverter(documents.ElementVisitor): return html_paths.empty - def visit_note_reference(self, note_reference): + def visit_note_reference(self, note_reference, context): self._note_references.append(note_reference) note_number = len(self._note_references) return [ @@ -196,8 +230,8 @@ class _DocumentConverter(documents.ElementVisitor): ] - def visit_note(self, note): - note_body = self._visit_all(note.body) + [ + def visit_note(self, note, context): + note_body = self._visit_all(note.body, context) + [ html.collapsible_element("p", {}, [ html.text(" "), html.element("a", {"href": "#" + self._note_ref_html_id(note)}, [ @@ -210,7 +244,7 @@ class _DocumentConverter(documents.ElementVisitor): ] - def visit_comment_reference(self, reference): + def visit_comment_reference(self, reference, context): def nodes(): comment = self._comments[reference.comment_id] count = len(self._referenced_comments) + 1 @@ -232,10 +266,10 @@ class _DocumentConverter(documents.ElementVisitor): return html_path.wrap(nodes) - def visit_comment(self, referenced_comment): + def visit_comment(self, referenced_comment, context): label, comment = referenced_comment # TODO remove duplication with notes - body = self._visit_all(comment.body) + [ + body = self._visit_all(comment.body, context) + [ html.collapsible_element("p", {}, [ html.text(" "), html.element("a", {"href": "#" + self._reference_html_id("comment", comment.comment_id)}, [ @@ -253,8 +287,12 @@ class _DocumentConverter(documents.ElementVisitor): ] - def _visit_all(self, elements): - return lists.flat_map(self.visit, elements) + def _visit_all(self, elements, context): + return [ + html_node + for element in elements + for html_node in self.visit(element, context) + ] def _find_html_path_for_paragraph(self, paragraph): diff --git a/mammoth/documents.py b/mammoth/documents.py index dbd78dc..b440846 100644 --- a/mammoth/documents.py +++ b/mammoth/documents.py @@ -216,4 +216,5 @@ class CommentReference(Element): comment_reference = CommentReference -ElementVisitor = cobble.visitor(Element) +def element_visitor(args): + return cobble.visitor(Element, args=args) diff --git a/mammoth/lists.py b/mammoth/lists.py index db1f9ca..28468fe 100644 --- a/mammoth/lists.py +++ b/mammoth/lists.py @@ -23,6 +23,12 @@ def flat_map(func, values): ] +def find_index(predicate, values): + for index, value in enumerate(values): + if predicate(value): + return index + + if sys.version_info[0] == 2: map = map filter = filter @@ -9,7 +9,7 @@ def read(fname): _install_requires = [ - "cobble>=0.1.2,<0.2", + "cobble>=0.1.3,<0.2", ] if sys.version_info[:2] <= (2, 6): diff --git a/tests/conversion_tests.py b/tests/conversion_tests.py index 3dfb9f3..3fd9e34 100644 --- a/tests/conversion_tests.py +++ b/tests/conversion_tests.py @@ -317,6 +317,35 @@ def docx_table_is_converted_to_table_in_html(): @istest +def header_rows_are_wrapped_in_thead(): + table = documents.table([ + documents.table_row([documents.table_cell([])], is_header=True), + documents.table_row([documents.table_cell([])], is_header=True), + documents.table_row([documents.table_cell([])], is_header=False), + ]) + result = convert_document_element_to_html(table) + expected_html = ( + "<table>" + + "<thead><tr><th></th></tr><tr><th></th></tr></thead>" + + "<tbody><tr><td></td></tr></tbody>" + + "</table>") + assert_equal(expected_html, result.value) + + +@istest +def tbody_is_omitted_if_all_rows_are_headers(): + table = documents.table([ + documents.table_row([documents.table_cell([])], is_header=True), + ]) + result = convert_document_element_to_html(table) + expected_html = ( + "<table>" + + "<thead><tr><th></th></tr></thead>" + + "</table>") + assert_equal(expected_html, result.value) + + +@istest def empty_cells_are_preserved_in_table(): table = documents.table([ documents.table_row([ |