Unverified Commit b8446a3e authored by Maxime Lefrançois's avatar Maxime Lefrançois
Browse files

better management of Figure identifiers in SAREF4LIFT

parent a61ac989

src/saref_pypeline/docgen/ts2md_extractor.py

+9 −2
Original line number Diff line number Diff line
@@ -853,7 +853,7 @@ class TS2MDExtractor: 
        """Figure title"""

        md = self.extract_inner_content(paragraph, extract_format)



        match = re.match(r"(\w+[\s\xa0]+[A-Z0-9\.]+)[: \xa0]*(.*)", md)

        match = re.match(r"(\w+[\s\xa0]+[A-Z0-9\.-]+)[: \xa0]*(.*)", md)

        if not match:

            return ""

        label, caption = match.group(1), match.group(2)
@@ -875,6 +875,13 @@ class TS2MDExtractor: 
        self, paragraph: Paragraph, extract_format: ExtractFormat = ExtractFormat.MD

    ):

        """Figure layout, do nothing"""

        if paragraph.text.startswith("Figure"):

            logger.warning(f"Wrong style for {paragraph.text}")

            return self.extract_TF(paragraph, extract_format)

        elif paragraph.text.strip():

            logger.warning(f"Styl FL should only be for Figure layout. Got {paragraph.text}")

            return self.extract_Normal(paragraph, extract_format)

        else:

            return ""



    def extract_NF(

src/saref_pypeline/docgen/ts_generator.py

+7 −7
Original line number Diff line number Diff line
@@ -935,8 +935,6 @@ class TSGenerator: 
            )

            return



        for p in soup.find_all("p"):

            p.attrs["class"] = "Normal"

        for li in soup.find_all("li"):

            li.string = re.sub(r" *: *", "\t", li.string, 1)

            li["data-docx-pstyle"] = P_STYLE.EW
@@ -957,8 +955,6 @@ class TSGenerator: 
            )

            return



        for p in soup.find_all("p"):

            p.attrs["class"] = "Normal"

        for li in soup.find_all("li"):

            for desc in li.descendants:

                if isinstance(desc, NavigableString):
@@ -966,7 +962,11 @@ class TSGenerator: 
                    if n:  # replaced once

                        desc.replace_with(new_text)

                        break



            if li.find_next_sibling("li") is not None:

                li["data-docx-pstyle"] = P_STYLE.EW

            else:

                li["data-docx-pstyle"] = P_STYLE.EX

        self.insert_soup(soup)



    def edit_history(self):
@@ -1039,7 +1039,7 @@ class TSGenerator: 
    # Methods for inserting soup

    # ---------------------------------------------------------------------



    @lru_cache

    # @lru_cache

    def get_soup(self, file_name: str) -> BeautifulSoup | None:

        """

        Load a file (HTML or Markdown) as a BeautifulSoup object.
@@ -1421,7 +1421,7 @@ class TSGenerator: 
        self.new_paragraph(style=P_STYLE.TF)

        if el.a:

            el.a.replace_with(el.a.text)

        match = re.match(r"^Figure (\d+)", el.get_text())

        match = re.match(r"^Figure ([\d\.-]+)", el.get_text())

        id = f"Figure_{match.group(1)}" if match else None

        with Bookmark(self, id):

            self.insert_soup_children(el)