Update add_xmlid_to_divs.ipynb

%% Cell type:markdown id: tags:

# Adding semi-automatic @xml:id to structured TEI-XML files

Script constructing identifiers for TEI-XML divs, according to the ConDÉ project schema, all values separated by `-`, all body div numbers formatted with three digits and all front|back div numbers formatted with two digits.

For `//tei:text/tei:front` and `//tei:text/tei:back` divs, the construction is as follows:

* source id,

* type of edition (base/txm/simplified),

* current version (alpha/beta)

* frontMatter or backMatter

* number of current front/div or back/div

* subtype of current front/div if any,

* number of current front/div/div if subject div is inside a div itself

⋅⋅* (max 2 levels of div in front and back).

* number of current front/div/div if subject div is inside a div itself (max 2 levels of div in front and back).

For `//tei:text/tei:body` divs, construction is as follows:

* source id,

* type of edition (base/txm/simplified),

* current version (alpha/beta)

* current part div number,

* current chapter div number

⋅⋅* (if subject div is a chapter or section)

* current section div number

⋅⋅* (if subject div is a section).

* current chapter div number (if subject div is a chapter or section)

* current section div number (if subject div is a section).

All body divs need to be typed (`part`/`chapter`/`section`) for this script to function.

### Imports and declarations

%% Cell type:code id: tags:

``` python

import xml.etree.ElementTree as ET

ET.register_namespace("", "http://www.tei-c.org/ns/1.0")

ET.register_namespace('xml','http://www.w3.org/XML/1998/namespace')

```

%% Cell type:markdown id: tags:

### FUNCTION: contains all the actual code of this file

%% Cell type:code id: tags:

``` python

def add_ids(xml_in, xml_out, fileID):

    """

    Function taking one TEI-XML file and adding @xml:id to <tei:div> elements.

    :param xml_in: The local path to the TEI-XML file needing <tei:div> identification, as a string.

    :param xml_out: The local path to the new TEI-XML file with identified <tei:div>, as a string.

    :param fileID: A string to prefix all identifiers, at best with only small letters inside.

                    Meant to use the corpus identifier of the current source.

    """

    # Starting a new file, we make counters for each type of div.

    section_counter = 0

    chapter_counter = 0

    part_counter = 0

    front_counter = 0

    back_counter = 0

    # Open and parse current TEI-XML file.

    tree = ET.parse(xml_in)

    root = tree.getroot()

    # Add the specified source ID to the <tei:text> element.

    textElement = root.find('.//{http://www.tei-c.org/ns/1.0}text')

    textElement.set('{http://www.w3.org/XML/1998/namespace}id', fileID)

    # START WITH THE FRONT MATTER DIVs.

    for item in root.findall(".//{http://www.tei-c.org/ns/1.0}front/*"):

    #for item in root.findall(".//front/*"):

        # Found one more div: add 1 to counter.

        front_counter += 1

        # If current element is a <tei:titlePage> element,

        # this will be included in its identifier.

        if item.tag == "{http://www.tei-c.org/ns/1.0}titlePage":

        #if item.tag == "titlePage":

            frontID = fileID + "-frontMatter-" + str("{:02}".format(front_counter)) + "-titlepage"

            item.set("{http://www.w3.org/XML/1998/namespace}id", frontID)

        # If current element is not a <tei:titlePage> element,

        # and it has an @type, this will be included in its identifier.

        elif item.get("type"):

            frontID = fileID + "-frontMatter-" + str("{:02}".format(front_counter)) + "-" + item.get("type")

            item.set("{http://www.w3.org/XML/1998/namespace}id", frontID)

        # If current element is not a <tei:titlePage> element,

        # and it has no @type, its identifier will only have its number.

        else:

            frontID = fileID + "-frontMatter-" + str("{:02}".format(front_counter))

            item.set("{http://www.w3.org/XML/1998/namespace}id", frontID)

    # THEN DO THE BACK MATTER DIVs.

    for item in root.findall(".//{http://www.tei-c.org/ns/1.0}back/*"):

    #for item in root.findall(".//back/*"):

        # Found one more div: add 1 to counter.

        back_counter += 1

        # If current element has an @type, this will be included in its identifier.

        if item.get("subtype"):

            backID = fileID + "-backMatter-" + str("{:02}".format(back_counter)) + "-" + item.get("subtype")

            item.set("{http://www.w3.org/XML/1998/namespace}id", backID)

        # Otherwise, we will use its number only.

        else:

            backID = fileID + "-backMatter-" + str("{:02}".format(back_counter))

            item.set("{http://www.w3.org/XML/1998/namespace}id", backID)

    # FINALLY, DO THE MAIN CONTENT DIVs.

    #for part in root.findall(".//{http://www.tei-c.org/ns/1.0}div[@type='part']"):

    for part in root.findall(".//{http://www.tei-c.org/ns/1.0}body/{http://www.tei-c.org/ns/1.0}div[@type='part']"):

        # When entering a new part div, start chapter numbers anew, add 1

        # to part counter and make the div id.

        chapter_counter = 0

        part_counter += 1

        part_identifier = fileID + "-" + str("{:03}".format(part_counter))

        # If there is no @xml:id, make it. Otherwise, replace it.

        if part.get("{http://www.w3.org/XML/1998/namespace}id"):

            del part.attrib['{http://www.w3.org/XML/1998/namespace}id']

        part.set("{http://www.w3.org/XML/1998/namespace}id", part_identifier)

        #for chapter in part.findall(".//{http://www.tei-c.org/ns/1.0}div[@type='chapter']"):

        for chapter in part.findall(".//{http://www.tei-c.org/ns/1.0}div[@type='chapter']"):

            # When entering a new chapter div, start section numbers anew, add 1

            # to chapter counter and make the div id.

            section_counter = 0

            chapter_counter += 1

            chapter_identifier = fileID + "-" + str("{:03}".format(part_counter)) + "-" + str("{:03}".format(chapter_counter))

            # If there is no @xml:id, make it. Otherwise, replace it.

            if chapter.get("{http://www.w3.org/XML/1998/namespace}id"):

                del chapter.attrib["{http://www.w3.org/XML/1998/namespace}id"]

            chapter.set("{http://www.w3.org/XML/1998/namespace}id", chapter_identifier)

            #for section in chapter.findall(".//{http://www.tei-c.org/ns/1.0}div[@type='section']"):

            for section in chapter.findall(".//{http://www.tei-c.org/ns/1.0}div[@type='section']"):

                section_counter += 1

                section_identifier = fileID + "-" + str("{:03}".format(part_counter)) + "-" + str("{:03}".format(chapter_counter)) + "-" + str("{:03}".format(section_counter))

                if section.get("{http://www.w3.org/XML/1998/namespace}id"):

                    del section.attrib["{http://www.w3.org/XML/1998/namespace}id"]

                section.set("{http://www.w3.org/XML/1998/namespace}id", section_identifier)

    # Write final TEI-XML file into specified output path.

    tree.write(xml_out, encoding="unicode")

```

%% Cell type:markdown id: tags:

### Apply the function to desired files.

Here is where you put Python to work and change settings with the function parameters:

* a string containing the path of the input file,

* a string containing the path for an output file,

* a prefix for all `<tei:div>` identifiers in the file.

%% Cell type:code id: tags:

``` python

add_ids(

    "/home/erminea/Documents/CONDE/nov-21_renum/terrien_base.xml",

    "/home/erminea/Documents/CONDE/nov-21_divID/terrien_base.xml",

    "terrien-base-beta"

)

```