o jE@sdZddlmZddlmZgdZddlmZddlm Z ddl m Z dd l m Z m Z mZmZmZmZmZmZdd lmZdd lmZdd lmZd#ddZ   d$d%ddZ    d&d'd!d"ZdS)(z.This is a python implementation of wcswidth().) annotations)Optional)zwcwidth._constantszwcwidth._wcwidthzwcwidth.bisearchzwcwidth.table_graphemezwcwidth.table_vs16)table_grapheme_overrideswcwidth)bisearch)_EMOJI_ZWJ_SET_ISC_VIRAMA_SET_CATEGORY_MC_TABLE_FITZPATRICK_RANGE_REGIONAL_INDICATOR_SETresolve_terminalget_term_overrides_clamp_ambiguous_width)VS15_WIDE_TO_NARROW)VS16_NARROW_TO_WIDE)GRAPHEME_EXTENDtextstrstartintendreturncCs|d}||kr"tt||tr"|d7}||kr"tt||ts||krgt||dkr1 |S|d7}||krdt||tvrd|d7}||krctt||trc|d7}||krctt||tsRq" |S|S)z Scan forward from *start* (base character) to end of a ZWJ grapheme cluster. Follows the UAX #29 GB11 pattern (ExtPict Extend* ZWJ x ExtPict) chained repeatedly until no more ZWJ joins are found. r )rordrr )rrridxr/root/aizidognhua/tmp/workspace/projects/ec89d86c-575f-41c9-af57-ac45cbdbf775/venv/lib/python3.10/site-packages/wcwidth/_wcswidth.py_scan_zwj_cluster_ends" rNautopwcsn Optional[int]unicode_versionambiguous_widthcs|dur|r|rt|Stdkrtnfdd}|dur(t|nt|t|}d}d}d}d} d} d} d} td } td }t}||krK||}t |}|d kru| r_|d7}n|d|krnd} d} |d 7}nd} |d7}qG|d kr|dkr|| | rd } d}|d7}qG|d kr|dkrt| |r| d kr|d8}d}|d7}qG|dkr|t vrd}|d}|dkrt ||t vr|d7}|d8}|dkrt ||t vs|d dkr|} |d7}qGnt d|krt dkrnn | t vr|d7}qG||}|dkrdS|dkr'| rd } n | r|| 7}|} n|} |}|} |} d} n|t vr/d} n|dkrA||trAd } d}d} nd} |d7}||ksL| rR|| 7}|S)a Given a unicode string, return its printable length on a terminal. See :ref:`Specification` for details of cell measurement. This implementation differs from Markus Khun's original POSIX C implementation, in that this ``wcswidth()`` processes graphemes strings yielded by :func:`wcwidth.iter_graphemes` defined by `Unicode Standard Annex #29`_. POSIX wcswidth(3) is not grapheme-aware and does not measure many kinds of Emojis or complex scripts correctly. :param pwcs: Measure width of given unicode string. :param n: When ``n`` is None (default), return the length of the entire string, otherwise only the first ``n`` characters are measured. :param unicode_version: Ignored. Retained for backwards compatibility. .. deprecated:: 0.3.0 Only the latest Unicode version is now shipped. :param ambiguous_width: Width to use for East Asian Ambiguous (A) characters. Default is ``1`` (narrow). Set to ``2`` for CJK contexts. :returns: The width, in cells, needed to display the first ``n`` characters of the unicode string ``pwcs``. Returns ``-1`` for C0 and C1 control characters! .. _`Unicode Standard Annex #29`: https://www.unicode.org/reports/tr29/ Nrc t|dSNr rcr%rrf zwcswidth..rF9.0.0rT)isascii isprintablelenrrminrrrrr r r r r )r!r"r$r%_wcwidthr total_widthrlast_measured_idxlast_measured_ucslast_measured_wprev_was_virama cluster_width vs16_nw_table vs15_wn_table _bisearchcharucs ri_beforejwrr*rwcswidth=s$          QrGT term_program bool | strc)s|dur|r|rt|Stt|}|r6t|}|j}|j}|j} |j } |j } |j } t |} nd}d}d} d} d} d} i} dkrJtnfdd}|durXt|nt|t|}d}d}d}d}d}d }d}d}d}td }td }t}||krb||}t|}|d kr|r|d7}nG|d|kr| r|dkr|tvrt|||}|||}| |} | dur|| |7}d}d}d}d }d}|}q{d}d }|d 7}nd }|d7}q{|d kr|dkr|r|||rn|||rd }d}|d7}q{|dkr(|dkr(|}!t|!|}"| rt|!| rd }"|"r!|d kr!|d8}d}|d7}q{|dkr|tvrkd}#|d}$|$dkr\t||$tvr\|#d7}#|$d8}$|$dkr\t||$tvsF|#d dkrj|}|d7}q{ntd|kr{tdkrnn |tvr|d7}q{||}%|%dkrdS|%d kr|rt||rd}%n|%d kr| rt|| rd}%|%dkr| rt|| rd }%n|%dkr| rt|| rd}%|%dkr>|rd }nV|r/d }&| r!|dkr!|||d}'| |'} | dur|| }d}&d}n|||}(| |(} | dur|| }n ||7}n||7}|&s.|%}|}|}n|%}|}|}|}|}|%}d }n|tvrFd}n|dkrX||trXd }d}d }nd }|d7}||ks|r| r|dkr|||}(| |(} | dur|| }|S||7}|S||7}|S)a Given a unicode string, return its printable length on a terminal given by ``term_program``. See :ref:`Specification` for details of cell measurement. Unlike :func:`wcswidth`, this function applies per-terminal correction tables for emoji presentation and grapheme clusters. :param pwcs: Measure width of given unicode string. :param n: When ``n`` is None (default), return the length of the entire string, otherwise only the first ``n`` characters are measured. :param unicode_version: Ignored. Retained for backwards compatibility. :param ambiguous_width: Width to use for East Asian Ambiguous (A) characters. Default is ``1`` (narrow). Set to ``2`` for CJK contexts. :param term_program: Terminal software identifier for table correction. ``True`` (default) reads the ``TERM_PROGRAM`` or ``TERM`` environment variable for auto-detection. ``False`` disables override lookup. Accepts a canonical terminal name matching :func:`list_term_programs`, such as from XTVERSION_, ENQ_, or ``TERM_PROGRAM``. .. versionadded:: 0.8.0 :returns: The width, in cells, needed to display the first ``n`` characters of the unicode string ``pwcs``. Returns ``-1`` for C0 and C1 control characters! Nrrcr&r'rr(r*rrr+ r,zwcstwidth..rr-r.Fr/rr0r1r2r3T)r4r5r6rrrnarrower vs16_narrower vs15_widerzeroer narrow_wider narrow_zeroerrgetrr7rrrrr rr r r r ))r!r"r$r%rHterm_canonical overrides _narrower_vs16_narrower _vs15_wider_zeroer _narrow_wider_narrow_zeroer_grapheme_overridesr8rr9rr:r;r<r= cluster_starttotal_before_clusterr>r?r@rArBrC cluster_endcluster override_wbase_ucs vs15_narrowrDrErFflushed candidate cluster_textrr*r wcstwidths@%              $              rd)rrrrrrrr)Nr r) r!rr"r#r$rr%rrr)Nr rT) r!rr"r#r$rr%rrHrIrr)__doc__ __future__rtypingr__lazy_modules__rr8rr _constantsr r r r r rrr table_vs15r table_vs16rtable_graphemerrrGrdrrrrs,     (