Skip to content

parsing

The internals of the TIM engine.

ContextDict

Bases: TypedDict

A dictionary to hold context about a markup language's environment.

It has two sub-dicts:

  • aliases
  • macros

For information about what they do and contain, see the MarkupLanguage docs.

Source code in pytermgui/markup/parsing.py
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
class ContextDict(TypedDict):
    """A dictionary to hold context about a markup language's environment.

    It has two sub-dicts:

    - aliases
    - macros

    For information about what they do and contain, see the
    [MarkupLanguage docs](/reference/pytermgui/markup/
    language#pytermgui.markup.language.MarkupLanguage).
    """

    aliases: dict[str, str]
    macros: dict[str, MacroType]

MacroType

Bases: Protocol

A protocol for TIM macros.

Source code in pytermgui/markup/parsing.py
49
50
51
52
53
54
55
class MacroType(Protocol):  # pylint: disable=too-few-public-methods
    """A protocol for TIM macros."""

    def __call__(  # pylint: disable=no-method-argument, no-self-argument
        *args: str,
    ) -> str:
        """Applies the macro."""

__call__(args)

Applies the macro.

Source code in pytermgui/markup/parsing.py
52
53
54
55
def __call__(  # pylint: disable=no-method-argument, no-self-argument
    *args: str,
) -> str:
    """Applies the macro."""

consume_tag(tag)

Consumes a tag text, returns the associated Token.

Source code in pytermgui/markup/parsing.py
 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
def consume_tag(tag: str) -> Token:  # pylint: disable=too-many-return-statements
    """Consumes a tag text, returns the associated Token."""

    if tag in STYLES:
        return StyleToken(tag)

    if tag.startswith("/"):
        return ClearToken(tag)

    if tag.startswith("!"):
        matchobj = RE_MACRO.match(tag)

        if matchobj is not None:
            name, args = matchobj.groups()

            if args is None:
                return MacroToken(name, tuple())

            return MacroToken(name, tuple(args.split(":")))

    if tag.startswith("~"):
        return HLinkToken(tag[1:])

    if tag.startswith("(") and tag.endswith(")"):
        values = tag[1:-1].split(";")
        if len(values) != 2:
            raise MarkupSyntaxError(
                tag,
                f"should have exactly 2 values separated by `;`, not {len(values)}",
                "",
            )

        return CursorToken(tag[1:-1], *map(int, values))

    if tag == "#auto":
        return PseudoToken("#auto")

    token: Token
    try:
        token = ColorToken(tag, Color.parse(tag, localize=False))

    except ColorSyntaxError:
        token = AliasToken(tag)

    finally:
        return token  # pylint: disable=lost-exception

create_context_dict()

Creates a new context dictionary, initializing its sub-dicts.

Returns:

Type Description
ContextDict

A dictionary with aliases and macros defined as empty sub-dicts.

Source code in pytermgui/markup/parsing.py
75
76
77
78
79
80
81
82
def create_context_dict() -> ContextDict:
    """Creates a new context dictionary, initializing its sub-dicts.

    Returns:
        A dictionary with `aliases` and `macros` defined as empty sub-dicts.
    """

    return {"aliases": {}, "macros": {}}

eval_alias(text, context)

Evaluates a space-delimited string of alias tags into their underlying value.

Parameters:

Name Type Description Default
text str

A space-separated string containing the aliases.

required

Returns:

Type Description
str

The space-separated string that the input aliases represent.

Source code in pytermgui/markup/parsing.py
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
def eval_alias(text: str, context: ContextDict) -> str:
    """Evaluates a space-delimited string of alias tags into their underlying value.

    Args:
        text: A space-separated string containing the aliases.

    Returns:
        The space-separated string that the input aliases represent.
    """

    aliases = context["aliases"]

    evaluated = ""
    for tag in text.split():
        if tag not in aliases:
            evaluated += tag + " "
            continue

        evaluated += eval_alias(aliases[tag], context) + " "

    return evaluated.rstrip(" ")

get_markup(text)

Gets the markup representing an ANSI-coded string.

Source code in pytermgui/markup/parsing.py
527
528
529
530
def get_markup(text: str) -> str:
    """Gets the markup representing an ANSI-coded string."""

    return tokens_to_markup(list(tokenize_ansi(text)))

optimize_markup(markup)

Optimizes markup by tokenizing it, optimizing the tokens and converting it back to markup.

Source code in pytermgui/markup/parsing.py
533
534
535
536
def optimize_markup(markup: str) -> str:
    """Optimizes markup by tokenizing it, optimizing the tokens and converting it back to markup."""

    return tokens_to_markup(list(optimize_tokens(list(tokenize_markup(markup)))))

optimize_tokens(tokens)

Optimizes a stream of tokens, only yielding functionally relevant ones.

Parameters:

Name Type Description Default
tokens list[Token]

Any list of Token objects. Usually obtained from tokenize_markup or tokenize_ansi.

required

Yields:

Type Description
Iterator[Token]

All those tokens within the input iterator that are functionally relevant, keeping their order.

Source code in pytermgui/markup/parsing.py
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
def optimize_tokens(tokens: list[Token]) -> Iterator[Token]:
    """Optimizes a stream of tokens, only yielding functionally relevant ones.

    Args:
        tokens: Any list of Token objects. Usually obtained from `tokenize_markup`
            or `tokenize_ansi`.

    Yields:
        All those tokens within the input iterator that are functionally relevant,
            keeping their order.
    """

    previous: list[Token] = []
    current_tag_group: list[Token] = []

    def _diff_previous() -> Iterator[Token]:
        """Find difference from the previously active list of tokens."""

        applied = previous.copy()

        for tkn in current_tag_group:
            targets = []

            clearer = Token.is_clear(tkn)
            if Token.is_clear(tkn):
                targets = [tkn.targets(tag) for tag in applied]

            if tkn in previous and not clearer:
                continue

            if clearer and not any(targets):
                continue

            applied.append(tkn)
            yield tkn

    def _remove_redundant_color(token: Token) -> None:
        """Removes non-functional colors.

        These happen in the following ways:
        - Multiple colors of the same channel (fg/bg) are present.
        - A color is applied, then a clearer clears it.
        """

        for applied in current_tag_group.copy():
            if Token.is_clear(applied) and applied.targets(token):
                current_tag_group.remove(applied)

            if not Token.is_color(applied):
                continue

            old = applied.color

            if old.background == new.background:
                current_tag_group.remove(applied)

    for token in tokens:
        if Token.is_plain(token):
            yield from _diff_previous()
            yield token

            previous = current_tag_group.copy()

            continue

        if Token.is_color(token):
            new = token.color

            _remove_redundant_color(token)

            if not any(token.markup == applied.markup for applied in current_tag_group):
                current_tag_group.append(token)

            continue

        if token.is_style():
            if not any(token == tag for tag in current_tag_group):
                current_tag_group.append(token)

            continue

        if Token.is_clear(token):
            applied = False
            for tag in current_tag_group.copy():
                if token.targets(tag) or token == tag:
                    current_tag_group.remove(tag)
                    applied = True

            if not applied:
                continue

        current_tag_group.append(token)

    yield from _diff_previous()

parse(text, optimize=False, context=None, append_reset=True, ignore_unknown_tags=True)

Parses markup into the ANSI-coded string it represents.

Parameters:

Name Type Description Default
text str

Any valid markup.

required
optimize bool

If set, optimize_tokens will optimize the tokens found within the input markup before usage. This will incur a (minor) performance hit.

False
context ContextDict | None

The context that aliases and macros found within the markup will be searched in.

None
append_reset bool

If set, [/] will be appended to the token iterator, clearing all styles.

True
ignore_unknown_tags bool

If set, the MarkupSyntaxError coming from unknown tags will be silenced.

True

Returns:

Type Description
str

The ANSI-coded string that the markup represents.

Source code in pytermgui/markup/parsing.py
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
def parse(
    text: str,
    optimize: bool = False,
    context: ContextDict | None = None,
    append_reset: bool = True,
    ignore_unknown_tags: bool = True,
) -> str:
    """Parses markup into the ANSI-coded string it represents.

    Args:
        text: Any valid markup.
        optimize: If set, `optimize_tokens` will optimize the tokens found within the
            input markup before usage. This will incur a (minor) performance hit.
        context: The context that aliases and macros found within the markup will be
            searched in.
        append_reset: If set, `[/]` will be appended to the token iterator, clearing all
            styles.
        ignore_unknown_tags: If set, the `MarkupSyntaxError` coming from unknown tags
            will be silenced.

    Returns:
        The ANSI-coded string that the markup represents.
    """

    if context is None:
        context = create_context_dict()

    if append_reset and not text.endswith("/]"):
        text += "[/]"

    tokens = list(tokenize_markup(text))

    return parse_tokens(
        tokens,
        optimize=optimize,
        context=context,
        append_reset=append_reset,
        ignore_unknown_tags=ignore_unknown_tags,
    )

parse_alias(token, context, get_full)

Parses an alias token.

Source code in pytermgui/markup/parsing.py
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
def parse_alias(
    token: AliasToken, context: ContextDict, get_full: Callable[[], str]
) -> str:
    """Parses an alias token."""

    if token.value not in context["aliases"]:
        dump = json.dumps(context["aliases"], indent=2, default=str)

        raise MarkupSyntaxError(
            token.value, f"not defined in alias context: {dump}", get_full()
        )

    meaning = context["aliases"][token.value]

    return eval_alias(meaning, context).rstrip(" ")

parse_clear(token, _, get_full)

Parses a clearer token.

Source code in pytermgui/markup/parsing.py
380
381
382
383
384
385
386
387
388
389
def parse_clear(token: ClearToken, _: ContextDict, get_full: Callable[[], str]) -> str:
    """Parses a clearer token."""

    index = CLEARERS.get(token.value)
    if index is None:
        raise MarkupSyntaxError(
            token.value, "not a recognized clearer or alias", get_full()
        )

    return f"\x1b[{index}m"

parse_color(token, _, __)

Parses a color token.

Source code in pytermgui/markup/parsing.py
327
328
329
330
def parse_color(token: ColorToken, _: ContextDict, __: Callable[[], str]) -> str:
    """Parses a color token."""

    return token.color.get_localized().sequence

parse_cursor(token, _, __)

Parses a cursor token.

Source code in pytermgui/markup/parsing.py
392
393
394
395
396
397
def parse_cursor(token: CursorToken, _: ContextDict, __: Callable[[], str]) -> str:
    """Parses a cursor token."""

    ypos, xpos = map(lambda i: "" if i is None else i, token)

    return f"\x1b[{ypos};{xpos}H"

parse_macro(token, context, get_full)

Parses a macro token.

Returns:

Type Description
MacroType

A tuple containing the callable bound to the name, as well as the arguments

tuple[str, ...]

passed to it.

Source code in pytermgui/markup/parsing.py
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
def parse_macro(
    token: MacroToken, context: ContextDict, get_full: Callable[[], str]
) -> tuple[MacroType, tuple[str, ...]]:
    """Parses a macro token.

    Returns:
        A tuple containing the callable bound to the name, as well as the arguments
        passed to it.
    """

    func = context["macros"].get(token.value)

    if func is None:
        dump = json.dumps(context["macros"], indent=2, default=str)

        raise MarkupSyntaxError(
            token.value, f"not defined in macro context: {dump}", get_full()
        )

    return func, token.arguments

parse_plain(token, _, __)

Parses a plain token.

Source code in pytermgui/markup/parsing.py
321
322
323
324
def parse_plain(token: PlainToken, _: ContextDict, __: Callable[[], str]) -> str:
    """Parses a plain token."""

    return token.value

parse_style(token, _, __)

Parses a style token.

Source code in pytermgui/markup/parsing.py
333
334
335
336
337
338
def parse_style(token: StyleToken, _: ContextDict, __: Callable[[], str]) -> str:
    """Parses a style token."""

    index = STYLES[token.value]

    return f"\x1b[{index}m"

parse_tokens(tokens, *, optimize=False, context=None, append_reset=True, ignore_unknown_tags=True)

Parses a stream of tokens into the ANSI-coded string they represent.

Parameters:

Name Type Description Default
tokens list[Token]

Any list of Tokens, usually obtained from either tokenize_ansi or tokenize_markup.

required
optimize bool

If set, optimize_tokens will optimize the input iterator before usage. This will incur a (minor) performance hit.

False
context ContextDict | None

The context that aliases and macros found within the tokens will be searched in.

None
append_reset bool

If set, ClearToken("/") will be appended to the token iterator, clearing all styles.

True
ignore_unknown_tags bool

If set, the MarkupSyntaxError coming from unknown tags will be silenced.

True

Returns:

Type Description
str

The ANSI-coded string that the token stream represents.

Source code in pytermgui/markup/parsing.py
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
def parse_tokens(  # pylint: disable=too-many-branches, too-many-locals, too-many-statements
    tokens: list[Token],
    *,
    optimize: bool = False,
    context: ContextDict | None = None,
    append_reset: bool = True,
    ignore_unknown_tags: bool = True,
) -> str:
    """Parses a stream of tokens into the ANSI-coded string they represent.

    Args:
        tokens: Any list of Tokens, usually obtained from either `tokenize_ansi` or
            `tokenize_markup`.
        optimize: If set, `optimize_tokens` will optimize the input iterator before
            usage. This will incur a (minor) performance hit.
        context: The context that aliases and macros found within the tokens will be
            searched in.
        append_reset: If set, `ClearToken("/")` will be appended to the token iterator,
            clearing all styles.
        ignore_unknown_tags: If set, the `MarkupSyntaxError` coming from unknown tags
            will be silenced.

    Returns:
        The ANSI-coded string that the token stream represents.
    """

    if context is None:
        context = create_context_dict()

    token_list = _sub_aliases(tokens, context)

    # It's more computationally efficient to create this lambda once and reuse it
    # every time. There is no need to define a full function, as it just returns
    # a function return.
    get_full = (
        lambda: tokens_to_markup(  # pylint: disable=unnecessary-lambda-assignment
            token_list
        )
    )

    if optimize:
        token_list = list(optimize_tokens(token_list))

    if append_reset:
        token_list.append(ClearToken("/"))

    link = None
    output = ""
    segment = ""
    background = Color.parse("#000000")
    macros: list[MacroToken] = []
    unknown_aliases: list[Token] = []

    for token in token_list:
        if token.is_plain():
            value = _apply_macros(
                token.value, (parse_macro(macro, context, get_full) for macro in macros)
            )

            if len(unknown_aliases) > 0:
                output += f"[{' '.join(tkn.value for tkn in unknown_aliases)}]"
                unknown_aliases = []

            output += segment + (
                value if link is None else LINK_TEMPLATE.format(uri=link, label=value)
            )

            segment = ""
            continue

        if token.is_hyperlink():
            link = token.value
            continue

        if Token.is_macro(token):
            macros.append(token)
            continue

        if Token.is_clear(token):
            if token.value in ("/", "/~"):
                link = None

            found = False
            for macro in macros.copy():
                if token.targets(macro):
                    macros.remove(macro)
                    found = True
                    break

            if found and token.value != "/":
                continue

            if token.value.startswith("/!"):
                raise MarkupSyntaxError(
                    token.value, "has nothing to target", get_full()
                )

        if Token.is_color(token) and token.color.background:
            background = token.color

        if Token.is_pseudo(token):
            if token.value == "#auto":
                token = ColorToken("#auto", background.contrast)

        try:
            segment += PARSERS[type(token)](token, context, get_full)  # type: ignore

        except MarkupSyntaxError:
            if not ignore_unknown_tags:
                raise

            unknown_aliases.append(token)

    if len(unknown_aliases) > 0:
        output += f"[{' '.join(tkn.value for tkn in unknown_aliases)}]"

    output += segment

    return output

tokenize_ansi(text)

Converts some ANSI-coded text into a stream of tokens.

Parameters:

Name Type Description Default
text str

Any valid ANSI-coded text.

required

Yields:

Type Description
Iterator[Token]

The generated tokens, in the order they occur within the text.

Source code in pytermgui/markup/parsing.py
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
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
def tokenize_ansi(  # pylint: disable=too-many-locals, too-many-branches, too-many-statements
    text: str,
) -> Iterator[Token]:
    """Converts some ANSI-coded text into a stream of tokens.

    Args:
        text: Any valid ANSI-coded text.

    Yields:
        The generated tokens, in the order they occur within the text.
    """

    cursor = 0

    for matchobj in RE_ANSI.finditer(text):
        start, end = matchobj.span()

        csi = matchobj.groups()[0:2]
        link_osc = matchobj.groups()[2:4]

        if link_osc != (None, None):
            cursor = end
            uri, label = link_osc

            yield HLinkToken(uri)
            yield PlainToken(label)

            continue

        full, content = csi

        if cursor < start:
            yield PlainToken(text[cursor:start])

        cursor = end

        code = ""

        # Position
        posmatch = RE_POSITION.match(full)

        if posmatch is not None:
            ypos, xpos = posmatch.groups()
            if not ypos and not xpos:
                raise ValueError(
                    f"Cannot parse cursor when no position is supplied. Match: {posmatch!r}"
                )

            yield CursorToken(content, int(ypos) or None, int(xpos) or None)
            continue

        parts = content.split(";")

        state = None
        color_code = ""
        for part in parts:
            if state is None:
                if part in REVERSE_STYLES:
                    yield StyleToken(REVERSE_STYLES[part])
                    continue

                if part in REVERSE_CLEARERS:
                    yield ClearToken(REVERSE_CLEARERS[part])
                    continue

                if part in ("38", "48"):
                    state = "COLOR"
                    color_code += part + ";"
                    continue

                # standard colors
                try:
                    yield ColorToken(part, Color.parse(part, localize=False))
                    continue

                except ColorSyntaxError as exc:
                    raise ValueError(f"Could not parse color tag {part!r}.") from exc

            if state != "COLOR":
                continue

            color_code += part + ";"

            # Ignore incomplete RGB colors
            if (
                color_code.startswith(("38;2;", "48;2;"))
                and len(color_code.split(";")) != 6
            ):
                continue

            try:
                code = color_code

                if code.startswith(("38;2;", "48;2;", "38;5;", "48;5;")):
                    stripped = code[5:-1]

                    if code.startswith("4"):
                        stripped = "@" + stripped

                    code = stripped

                yield ColorToken(code, Color.parse(code, localize=False))

            except ColorSyntaxError:
                continue

            state = None
            color_code = ""

    remaining = text[cursor:]
    if len(remaining) > 0:
        yield PlainToken(remaining)

tokenize_markup(text)

Converts some markup text into a stream of tokens.

Parameters:

Name Type Description Default
text str

Any valid markup.

required

Yields:

Type Description
Iterator[Token]

The generated tokens, in the order they occur within the markup.

Source code in pytermgui/markup/parsing.py
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
def tokenize_markup(text: str) -> Iterator[Token]:
    """Converts some markup text into a stream of tokens.

    Args:
        text: Any valid markup.

    Yields:
        The generated tokens, in the order they occur within the markup.
    """

    cursor = 0
    length = len(text)
    has_inverse = False
    for matchobj in RE_MARKUP.finditer(text):
        full, escapes, content = matchobj.groups()
        start, end = matchobj.span()

        if cursor < start:
            yield PlainToken(text[cursor:start])

        if not escapes == "":
            _, remaining = divmod(len(escapes), 2)

            yield PlainToken(full[max(1 - remaining, 1) :])
            cursor = end

            continue

        for tag in content.split():
            if tag == "inverse":
                has_inverse = True

            if tag == "/inverse":
                has_inverse = False

            consumed = consume_tag(tag)
            if has_inverse:
                if consumed.markup == "/fg":
                    consumed = ClearToken("/fg")

                elif consumed.markup == "/bg":
                    consumed = ClearToken("/bg")

            yield consumed

        cursor = end

    if cursor < length:
        yield PlainToken(text[cursor:length])

tokens_to_markup(tokens)

Converts a token stream into the markup of its tokens.

Parameters:

Name Type Description Default
tokens list[Token]

Any list of Token objects. Usually obtained from tokenize_markup or tokenize_ansi.

required

Returns:

Type Description
str

The markup the given tokens represent.

Source code in pytermgui/markup/parsing.py
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
def tokens_to_markup(tokens: list[Token]) -> str:
    """Converts a token stream into the markup of its tokens.

    Args:
        tokens: Any list of Token objects. Usually obtained from `tokenize_markup` or
            `tokenize_ansi`.

    Returns:
        The markup the given tokens represent.
    """

    tags: list[Token] = []
    markup = ""

    for token in tokens:
        if token.is_plain():
            if len(tags) > 0:
                markup += f"[{' '.join(tag.markup for tag in tags)}]"

            markup += token.value
            tags = []

        else:
            tags.append(token)

    if len(tags) > 0:
        markup += f"[{' '.join(tag.markup for tag in tags)}]"

    return markup