Skip to content

cltk.exports

Scholar-facing export helpers.

doc_to_igt_html 

doc_to_igt_html(
    doc: Doc,
    *,
    include_translation: bool = True,
    include_gloss: bool = True,
    include_lemma: bool = False,
    include_morph: bool = False,
    max_sentences: Optional[int] = None
) -> str

Render a token-level IGT table in a self-contained HTML document.

Source code in cltk/exports/igt.py 
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
def doc_to_igt_html(
    doc: Doc,
    *,
    include_translation: bool = True,
    include_gloss: bool = True,
    include_lemma: bool = False,
    include_morph: bool = False,
    max_sentences: Optional[int] = None,
) -> str:
    """Render a token-level IGT table in a self-contained HTML document."""
    sentences = iter_sentences(doc, max_sentences)
    mode = _select_second_line(
        include_gloss=include_gloss,
        include_lemma=include_lemma,
        include_morph=include_morph,
    )
    blocks: list[str] = []
    for sent_idx, sentence in enumerate(sentences, start=1):
        words = iter_words(sentence)
        if not words:
            continue
        token_cells = "".join(
            f"<td>{html_escape(get_token_text(word))}</td>" for word in words
        )
        if mode == "morph":
            gloss_cells = "".join(
                f"<td>{html_escape(format_morph(word))}</td>" for word in words
            )
        elif mode == "lemma":
            gloss_cells = "".join(
                f"<td>{html_escape(get_lemma(word))}</td>" for word in words
            )
        elif mode == "gloss":
            gloss_cells = "".join(
                f"<td>{html_escape(get_gloss(word))}</td>" for word in words
            )
        else:
            gloss_cells = "".join(
                f"<td>{html_escape(get_token_text(word))}</td>" for word in words
            )
        translation_html = ""
        if include_translation:
            translation = _sentence_translation(doc, sent_idx - 1)
            if translation:
                translation_html = f'<div class="igt-translation">{html_escape(str(translation))}</div>'
        sentence_text_value = html_escape(sentence_text(doc, sentence, words))
        blocks.append(
            "\n".join(
                [
                    f'<section class="igt-sentence" id="igt-s{sent_idx}">',
                    f'<div class="igt-surface">{sentence_text_value}</div>',
                    '<table class="igt-table">',
                    f"<tr>{token_cells}</tr>",
                    f"<tr>{gloss_cells}</tr>",
                    "</table>",
                    translation_html,
                    "</section>",
                ]
            )
        )

    body = "\n".join(blocks)
    return (
        "\n".join(
            [
                "<!doctype html>",
                '<html lang="en">',
                "<head>",
                '<meta charset="utf-8" />',
                "<title>CLTK Token-level IGT</title>",
                "<style>",
                ".igt-sentence{margin:1.5rem 0;}",
                ".igt-surface{font-weight:600;margin-bottom:0.25rem;}",
                ".igt-table{border-collapse:collapse;width:100%;}",
                ".igt-table td{border:1px solid #ccc;padding:0.35rem;text-align:center;}",
                ".igt-translation{margin-top:0.4rem;font-style:italic;}",
                "</style>",
                "</head>",
                "<body>",
                "<h1>Token-level IGT</h1>",
                body,
                "</body>",
                "</html>",
            ]
        ).strip()
        + "\n"
    )

doc_to_igt_latex 

doc_to_igt_latex(
    doc: Doc,
    *,
    include_translation: bool = True,
    include_gloss: bool = True,
    include_lemma: bool = False,
    include_morph: bool = False,
    max_sentences: Optional[int] = None
) -> str

Render a token-level IGT table in LaTeX.

Gloss selection priority: 1) word.enrichment.gloss 2) first word.enrichment.lemma_translations entry 3) word.lemma 4) token string

Source code in cltk/exports/igt.py 
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
def doc_to_igt_latex(
    doc: Doc,
    *,
    include_translation: bool = True,
    include_gloss: bool = True,
    include_lemma: bool = False,
    include_morph: bool = False,
    max_sentences: Optional[int] = None,
) -> str:
    """Render a token-level IGT table in LaTeX.

    Gloss selection priority:
    1) ``word.enrichment.gloss``
    2) first ``word.enrichment.lemma_translations`` entry
    3) ``word.lemma``
    4) token string

    """
    sentences = iter_sentences(doc, max_sentences)
    mode = _select_second_line(
        include_gloss=include_gloss,
        include_lemma=include_lemma,
        include_morph=include_morph,
    )
    lines: list[str] = ["% Token-level IGT generated by CLTK", ""]
    for sent_idx, sentence in enumerate(sentences, start=1):
        words = iter_words(sentence)
        if not words:
            continue
        columns = " ".join(["l"] * len(words))
        lines.append(f"\\begin{{tabular}}{{{columns}}}")
        token_row = " & ".join(latex_escape(get_token_text(word)) for word in words)
        lines.append(f"{token_row} \\")
        if mode == "morph":
            gloss_row = " & ".join(latex_escape(format_morph(word)) for word in words)
        elif mode == "lemma":
            gloss_row = " & ".join(latex_escape(get_lemma(word)) for word in words)
        elif mode == "gloss":
            gloss_row = " & ".join(latex_escape(get_gloss(word)) for word in words)
        else:
            gloss_row = " & ".join(latex_escape(get_token_text(word)) for word in words)
        lines.append(f"{gloss_row} \\")
        if include_translation:
            translation = _sentence_translation(doc, sent_idx - 1)
            if translation:
                escaped = latex_escape(str(translation))
                lines.append(f"\\multicolumn{{{len(words)}}}{{l}}{{{escaped}}} \\")
        lines.append("\\end{tabular}")
        lines.append("")
    return "\n".join(lines).strip() + "\n"

doc_to_readers_guide_html 

doc_to_readers_guide_html(
    doc: Doc,
    *,
    title: Optional[str] = None,
    include_ipa: bool = True,
    include_gloss: bool = True,
    include_morph: bool = True,
    include_dependencies: bool = True,
    include_translation: bool = True,
    max_sentences: Optional[int] = None
) -> str

Render a self-contained HTML reader's guide for a Doc.

Source code in cltk/exports/readers_guide_html.py 
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
def doc_to_readers_guide_html(
    doc: Doc,
    *,
    title: Optional[str] = None,
    include_ipa: bool = True,
    include_gloss: bool = True,
    include_morph: bool = True,
    include_dependencies: bool = True,
    include_translation: bool = True,
    max_sentences: Optional[int] = None,
) -> str:
    """Render a self-contained HTML reader's guide for a ``Doc``."""
    sentences = iter_sentences(doc, max_sentences)
    doc_title = (
        title
        or (getattr(doc, "metadata", None) or {}).get("title")
        or (getattr(doc, "metadata", None) or {}).get("reference")
        or "Reader's Guide"
    )

    toc_items: list[str] = []
    sentence_blocks: list[str] = []
    for sent_idx, sentence in enumerate(sentences, start=1):
        words = iter_words(sentence)
        if not words:
            continue
        anchor = f"sentence-{sent_idx}"
        toc_items.append(f'<li><a href="#{anchor}">Sentence {sent_idx}</a></li>')
        sentence_value = html_escape(sentence_text(doc, sentence, words))
        translation_html = ""
        if include_translation:
            translation = _sentence_translation(doc, sent_idx - 1)
            if translation:
                translation_html = f'<div class="sentence-translation">{html_escape(str(translation))}</div>'

        token_strip_parts: list[str] = []
        for word in words:
            token = html_escape(get_token_text(word))
            gloss = get_gloss(word) if include_gloss else ""
            ipa_value, ipa_mode = get_ipa(word) if include_ipa else (None, None)
            tooltip_parts: list[str] = []
            if gloss:
                tooltip_parts.append(f"Gloss: {gloss}")
            if ipa_value:
                ipa_text = f"IPA: {ipa_value}"
                if ipa_mode:
                    ipa_text += f" ({ipa_mode})"
                tooltip_parts.append(ipa_text)
            tooltip_text = html_escape(" • ".join(tooltip_parts))
            data_attr = f' data-tooltip="{tooltip_text}"' if tooltip_text else ""
            token_strip_parts.append(f'<span class="token"{data_attr}>{token}</span>')
        token_strip = " ".join(token_strip_parts)

        card_blocks: list[str] = []
        for tok_idx, word in enumerate(words, start=1):
            token = html_escape(get_token_text(word))
            lemma = get_lemma(word)
            pos = get_upos_tag(word)
            feats = format_features(word)
            gloss = get_gloss(word) if include_gloss else ""
            ipa_value, ipa_mode = get_ipa(word) if include_ipa else (None, None)
            morph = format_morph(word) if include_morph else ""
            dep = getattr(word, "dependency_relation", None)
            dep_code = getattr(dep, "code", None) or getattr(dep, "tag", None)
            dep_name = getattr(dep, "name", None)
            governor = getattr(word, "governor", None)

            rows: list[str] = []
            if lemma:
                rows.append(
                    "".join(
                        [
                            "<dt>Lemma</dt>",
                            f'<dd>{html_escape(str(lemma))} <button class="copy" data-copy="{html_escape(str(lemma))}">Copy</button></dd>',
                        ]
                    )
                )
            if pos:
                rows.append(f"<dt>POS</dt><dd>{html_escape(str(pos))}</dd>")
            if include_morph and morph:
                rows.append(f"<dt>Morphology</dt><dd>{html_escape(morph)}</dd>")
            if include_gloss and gloss:
                rows.append(
                    "".join(
                        [
                            "<dt>Gloss</dt>",
                            f'<dd>{html_escape(gloss)} <button class="copy" data-copy="{html_escape(gloss)}">Copy</button></dd>',
                        ]
                    )
                )
            if include_ipa and ipa_value:
                ipa_label = html_escape(ipa_value)
                if ipa_mode:
                    ipa_label += (
                        f' <span class="muted">({html_escape(ipa_mode)})</span>'
                    )
                rows.append(f"<dt>IPA</dt><dd>{ipa_label}</dd>")
            if feats:
                rows.append(f"<dt>Features</dt><dd>{html_escape(feats)}</dd>")
            if include_dependencies and (dep_code or governor is not None):
                dep_label = dep_code or ""
                if dep_name and dep_name != dep_code:
                    dep_label = (
                        f"{dep_name} ({dep_code})" if dep_code else str(dep_name)
                    )
                governor_label = ""
                if governor is not None:
                    try:
                        head_index = int(governor)
                    except (TypeError, ValueError):
                        head_index = None
                    if head_index is not None and 0 <= head_index < len(words):
                        head_word = get_token_text(words[head_index])
                        governor_label = f"token {head_index + 1} ({head_word})"
                    elif head_index is not None:
                        governor_label = f"token {head_index + 1}"
                dep_details = " · ".join(
                    part for part in [dep_label, governor_label] if part
                )
                if dep_details:
                    rows.append(
                        f"<dt>Dependencies</dt><dd>{html_escape(dep_details)}</dd>"
                    )

            rows_html = "".join(rows) if rows else "<p>No details available.</p>"
            card_blocks.append(
                "\n".join(
                    [
                        '<details class="token-card">',
                        f"<summary>{token}</summary>",
                        f"<dl>{rows_html}</dl>",
                        "</details>",
                    ]
                )
            )

        sentence_blocks.append(
            "\n".join(
                [
                    f'<section id="{anchor}" class="sentence">',
                    f"<h2>Sentence {sent_idx}</h2>",
                    f'<div class="sentence-text">{sentence_value}</div>',
                    translation_html,
                    f'<div class="token-strip">{token_strip}</div>',
                    '<div class="token-cards">',
                    "\n".join(card_blocks),
                    "</div>",
                    "</section>",
                ]
            )
        )

    toc_html = ""
    if len(toc_items) > 1:
        toc_html = (
            '<nav class="toc"><h2>Contents</h2><ol>'
            + "".join(toc_items)
            + "</ol></nav>"
        )

    body = "\n".join(sentence_blocks)
    return (
        "\n".join(
            [
                "<!doctype html>",
                '<html lang="en">',
                "<head>",
                '<meta charset="utf-8" />',
                f"<title>{html_escape(str(doc_title))}</title>",
                "<style>",
                "body{font-family:Georgia,serif;line-height:1.6;margin:2rem;color:#1f1f1f;}",
                "h1,h2{font-family:'Palatino Linotype',serif;}",
                ".toc{background:#f5f5f5;padding:1rem;border-radius:8px;margin-bottom:1.5rem;}",
                ".sentence{margin:2rem 0;padding-bottom:1.5rem;border-bottom:1px solid #ddd;}",
                ".sentence-text{font-weight:600;margin-bottom:0.25rem;}",
                ".sentence-translation{font-style:italic;margin-bottom:0.75rem;color:#444;}",
                ".token-strip{display:flex;flex-wrap:wrap;gap:0.5rem;margin:0.75rem 0;}",
                ".token{position:relative;padding:0.2rem 0.4rem;border-radius:4px;background:#eef2ff;}",
                ".token[data-tooltip]:hover::after{content:attr(data-tooltip);position:absolute;left:0;top:100%;background:#111;color:#fff;padding:0.4rem;border-radius:4px;white-space:nowrap;font-size:0.8rem;z-index:5;}",
                ".token-cards{display:grid;grid-template-columns:repeat(auto-fit,minmax(220px,1fr));gap:0.75rem;}",
                ".token-card{background:#fafafa;border:1px solid #e2e2e2;border-radius:8px;padding:0.5rem;}",
                ".token-card summary{cursor:pointer;font-weight:600;margin-bottom:0.4rem;}",
                ".token-card dl{margin:0;}",
                ".token-card dt{font-weight:600;margin-top:0.3rem;}",
                ".token-card dd{margin:0 0 0.3rem 0.5rem;}",
                ".copy{margin-left:0.4rem;font-size:0.75rem;}",
                ".muted{color:#666;font-size:0.85rem;}",
                "</style>",
                "</head>",
                "<body>",
                f"<h1>{html_escape(str(doc_title))}</h1>",
                toc_html,
                body,
                "<script>",
                "document.addEventListener('click', (event) => {",
                "  const target = event.target;",
                "  if (target && target.classList.contains('copy')) {",
                "    const text = target.getAttribute('data-copy') || '';",
                "    if (navigator.clipboard) {",
                "      navigator.clipboard.writeText(text);",
                "    } else {",
                "      const temp = document.createElement('textarea');",
                "      temp.value = text;",
                "      document.body.appendChild(temp);",
                "      temp.select();",
                "      document.execCommand('copy');",
                "      document.body.removeChild(temp);",
                "    }",
                "  }",
                "});",
                "</script>",
                "</body>",
                "</html>",
            ]
        ).strip()
        + "\n"
    )

doc_to_tei_xml 

doc_to_tei_xml(
    doc: Doc,
    *,
    tei_minimal: bool = True,
    include_morph: bool = True,
    include_lemma: bool = True,
    include_gloss: bool = False,
    include_translation: bool = False,
    max_sentences: Optional[int] = None
) -> str

Render a TEI-ish XML document with standoff dependency relations.

Source code in cltk/exports/tei.py 
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
def doc_to_tei_xml(
    doc: Doc,
    *,
    tei_minimal: bool = True,
    include_morph: bool = True,
    include_lemma: bool = True,
    include_gloss: bool = False,
    include_translation: bool = False,
    max_sentences: Optional[int] = None,
) -> str:
    """Render a TEI-ish XML document with standoff dependency relations."""
    sentences = iter_sentences(doc, max_sentences)
    tei = ET.Element(f"{{{_TEI_NS}}}TEI")

    tei_header = ET.SubElement(tei, f"{{{_TEI_NS}}}teiHeader")
    file_desc = ET.SubElement(tei_header, f"{{{_TEI_NS}}}fileDesc")
    title_stmt = ET.SubElement(file_desc, f"{{{_TEI_NS}}}titleStmt")
    title = ET.SubElement(title_stmt, f"{{{_TEI_NS}}}title")
    metadata = getattr(doc, "metadata", None) or {}
    title.text = str(
        metadata.get("title") or metadata.get("reference") or "CLTK Document"
    )
    pub_stmt = ET.SubElement(file_desc, f"{{{_TEI_NS}}}publicationStmt")
    pub_p = ET.SubElement(pub_stmt, f"{{{_TEI_NS}}}p")
    pub_p.text = "Unpublished"
    source_desc = ET.SubElement(file_desc, f"{{{_TEI_NS}}}sourceDesc")
    source_p = ET.SubElement(source_desc, f"{{{_TEI_NS}}}p")
    source_p.text = "Generated by CLTK"

    lang_id, lang_name = _doc_language(doc)
    if lang_id or lang_name:
        profile_desc = ET.SubElement(tei_header, f"{{{_TEI_NS}}}profileDesc")
        lang_usage = ET.SubElement(profile_desc, f"{{{_TEI_NS}}}langUsage")
        language = ET.SubElement(lang_usage, f"{{{_TEI_NS}}}language")
        if lang_id:
            language.set("ident", lang_id)
        language.text = lang_name or ""

    app_version = _app_version()
    if app_version or not tei_minimal:
        encoding_desc = ET.SubElement(tei_header, f"{{{_TEI_NS}}}encodingDesc")
        app_info = ET.SubElement(encoding_desc, f"{{{_TEI_NS}}}appInfo")
        app = ET.SubElement(app_info, f"{{{_TEI_NS}}}application")
        app.set("ident", "cltk")
        if app_version:
            app.set("version", app_version)
        app.text = "CLTK"

    text = ET.SubElement(tei, f"{{{_TEI_NS}}}text")
    body = ET.SubElement(text, f"{{{_TEI_NS}}}body")
    div = ET.SubElement(body, f"{{{_TEI_NS}}}div")
    _set_attributes(div, [("type", "document")])
    div.set(f"{{{_XML_NS}}}id", _doc_id(doc))
    paragraph = ET.SubElement(div, f"{{{_TEI_NS}}}p")

    list_relation = None
    stand_off = ET.SubElement(tei, f"{{{_TEI_NS}}}standOff")
    list_relation = ET.SubElement(stand_off, f"{{{_TEI_NS}}}listRelation")
    list_relation.set("type", "dependency")

    for sent_num, sentence in enumerate(sentences, start=1):
        words = iter_words(sentence)
        if not words:
            continue
        s_elem = ET.SubElement(paragraph, f"{{{_TEI_NS}}}s")
        s_elem.set(f"{{{_XML_NS}}}id", make_sentence_id(sent_num))
        root_anchor = ET.SubElement(list_relation, f"{{{_TEI_NS}}}anchor")
        root_anchor.set(f"{{{_XML_NS}}}id", make_root_id(sent_num))

        for tok_num, word in enumerate(words, start=1):
            w_elem = ET.SubElement(s_elem, f"{{{_TEI_NS}}}w")
            attrs: list[tuple[str, str]] = []
            attrs.append((f"{{{_XML_NS}}}id", make_token_id(sent_num, tok_num)))
            attrs.append(("n", str(tok_num)))
            if include_lemma:
                lemma = get_lemma(word)
                if lemma:
                    attrs.append(("lemma", lemma))
            pos = get_upos_tag(word)
            if pos:
                attrs.append(("pos", pos))
            if include_morph:
                feats = format_features(word)
                if feats:
                    attrs.append(("msd", feats))
            if include_gloss:
                gloss = get_gloss(word)
                if gloss:
                    attrs.append(("gloss", gloss))
            _set_attributes(w_elem, attrs)
            w_elem.text = get_token_text(word)

        if include_translation:
            translation_map = getattr(doc, "sentence_translations", None) or {}
            translation = translation_map.get(sent_num - 1)
            text_value = getattr(translation, "text", None) if translation else None
            if text_value:
                note = ET.SubElement(s_elem, f"{{{_TEI_NS}}}note")
                note.set("type", "translation")
                note.text = str(text_value)

        for tok_num, word in enumerate(words, start=1):
            dep = getattr(word, "dependency_relation", None)
            dep_code = getattr(dep, "code", None) or getattr(dep, "tag", None)
            if not dep_code:
                continue
            governor = getattr(word, "governor", None)
            active = make_token_id(sent_num, tok_num)
            if governor is None:
                passive = make_root_id(sent_num)
            else:
                try:
                    head_index = int(governor)
                except (TypeError, ValueError):
                    continue
                if head_index < 0 or head_index >= len(words):
                    continue
                passive = make_token_id(sent_num, head_index + 1)
            rel = ET.SubElement(list_relation, f"{{{_TEI_NS}}}relation")
            rel_id = f"r-{active}-{passive}"
            _set_attributes(
                rel,
                [
                    (f"{{{_XML_NS}}}id", rel_id),
                    ("name", str(dep_code)),
                    ("active", f"#{active}"),
                    ("passive", f"#{passive}"),
                ],
            )

    _indent(tei)
    xml_bytes: bytes = ET.tostring(tei, encoding="utf-8", xml_declaration=True)
    return xml_bytes.decode("utf-8")

Submodules