xj UdZddlmZddlZddlZddlZddlmZddlm Z m Z m Z m Z ddl mZddlmZdd lmZmZdd lmZmZmZmZmZmZdd lmZdd lmZmZdd l m!Z!m"Z"m#Z#de$d<ej%&ddrdZ'n ddl(m)Z*e*Z'n #e+$rdZ'YnwxYwGddej,Z-Gdde Z. dGdHdZ/dId-Z0dJd2Z1dKd6Z2dLd;Z3dKd<Z4 dMd>d?dd@dAddBdCdNdFZ)dS)Oz*This is a python implementation of clip().) annotationsN)islice)LiteralCallableOptional NamedTuple)width)iter_graphemes) HyperlinkHyperlinkParams) _SGR_PATTERN_SGR_STATE_DEFAULT _SGRState_sgr_state_update_sgr_state_is_active_sgr_state_to_sequence)_clamp_ambiguous_width) TextSizingTextSizingParams)_SEQUENCE_CLASSIFY_HORIZONTAL_CURSOR_MOVEMENTINDETERMINATE_EFFECT_SEQUENCEz&Optional[Callable[..., Optional[str]]]_c_clipWCWIDTH_PYTHON)clipceZdZdZejZejZejZejZ dS)_HyperlinkActionz.Outcome of processing an OSC 8 hyperlink unit.N) __name__ __module__ __qualname____doc__enumautoNO_CLOSEEMPTYOUTSIDEVISIBLE>C:\Projects\Historical\venv\Lib\site-packages\wcwidth/_clip.pyrr)sH88ty{{H DIKKEdikkGdikkGGGr+rceZdZUdZded<dZded<dZded<dZd ed <dZd ed <dZ d ed <dZ ded <dZ ded<dS)_HyperlinkResultzn Result of processing an OSC 8 hyperlink. Only the fields relevant to each action are populated. ractionrint close_end inner_widthrstropen_seq clipped_inner close_seq clipped_width hl_col_endN) r r!r"r#__annotations__r1r2r4r5r6r7r8r*r+r,r.r.2s IKHMIMJr+r.resultr3captured_styleOptional[_SGRState] end_stylereturncf|.t|x}r||z}t||n|r|dz }|S)aZ Apply SGR prefix/suffix around *result*. If an SGR state was captured at the first visible character, prefix the result with the corresponding SGR sequence, and suffix with a reset if any styles remain active at the end of the clipped region. *end_style* is the style in effect after the final SGR sequence emitted within the clip window, or ``None`` when no such sequence was emitted, the style at the first visible character is still in effect. This matches :func:`wcwidth.propagate_sgr`, which decides the trailing reset by SGR state at the end of each line. Nz)rr)r:r;r=prefixs r,_apply_sgr_wraprACsP!+N;; ;6 %f_F )2C S S i F Mr+textstartr0endfillchartabsizeambiguous_width term_program bool | str control_codes$Literal['parse', 'strict', 'ignore']paramsr match_endcolc ^tj|| \} } | | fdkrttjS|| | } t | ||||}|dkrttj| S| |z}||ks| |krttj| |Std|| z }|| z }t| ||||||d| }t |||||}ttj | || || ||S) z Process an OSC 8 hyperlink unit. Finds the matching close sequence, measures the inner text width, and determines whether the hyperlink is empty, outside the clip window, or visible (requiring inner-text clipping). )rP)rJrFrGrHr)r1)r1r2F)rErFrGrH propagate_sgrrJ)r1r2r4r5r6r7r8) r find_closer.rr&r r'r(maxrr) make_open make_close)rBrCrDrErFrGrHrJrLrMrN close_startr1 inner_textr2r8inner_clip_startinner_clip_endr5r7s r,_process_hyperlinkrZYs*'1$ BBKY8++ 0 9:::i +,J-!K a 0 6)LLLL{"JUcSjj 0 8I,7999 91eck**3YN$n7'!# M]!M  !!#####%%#    r+cellsdict[int, tuple[str, int]] sequenceslist[tuple[int, int, str]]ci}|D]1\}}}||g||f2|D]} | |r!t |nd} |r!t |nd} t | | } g} d}t | |}||kr||dD]\}}| |||kr|dz }C||vr(||\}}| |||pdz }n*||cxkr| krnn| ||dz }||kt|D]+}||kr#||D]\}}| |,d | S)z Reconstruct the output string from painter's algorithm state. Walks columns left-to-right, interleaving escape sequences and cell content, filling gaps with *fillchar*. rPrr*r r) setdefaultappendvaluessortrSkeysmingetsortedjoin)r[r]rCrDrE seqs_by_colcol_posorderseq_textentries max_cell_col max_seq_colmax_colpartswalk_col col_limit_ cell_textcell_wcs r,_reconstruct_painterrxs'57K$-FF w++22E83DEEEE%%'' (-53uzz||$$$2L-8@#k&&(()))bK, ,,GEHGS!!I i  &??8R88 # #KAx LL " " " " s?? MH  u   %h Iv LL # # #  ! #HH0000L00000 X&&& MH! i  &K$$&& ' ''' y==*1~ ' ' 8 X&&&& 775>>r+rQbool4tuple[str, Optional[_SGRState], Optional[_SGRState]]c  |dk} g} d} d} d} d}|rtnd}| t|kr|| }| |kr,|dvr(| n|d| dz}|dkrn`|} N|dkrtj|| }|s| || dz } |dm|rk|it||}| /| |kr)| ||}|}  tj |x}rft||||||||||| }|j tjur<| ||} n|j tjur|j} n|j tjur| |jz } |j} n| |j| |j| |j|r| |} |rB|@t-j|jD]&}t||}|}'| |jz } |j} |d x}|d pd }|d }|Jt1t3j||||}||}| |kr<| |z|kr3| ||r| |} | |z } nN| |krC| |z|kr:g d fd }t;|| ||||||} |  |r| |} n| |z } |} |}| r)t?j|rtAd| d|| ||} *|dkrz|dkrH| || |zz z}| |kr6|| cxkr|krnn| d|r| |} | dz } | |k6n%|| cxkr|krnn| d| dz } tCtE|| }tG|||}|dkr&|| cxkr|krnn| |nz| |kr%| |z|kr| ||r| |} nO| |krI| |z|kr@| |tI|| |ztK|| z z|r| |} | |z } | t|z } | t|kd &| | |fS)z Clip text without cursor movement (simple append-to-output path). Returns ``(result, captured_style, end_style)``. The caller applies SGR wrapping. strictrN  r rP sgr_paramsrLrMrNts_metats_textrts_termrJsr3_wr0_colr>Nonec2|dSN)ra)rrrts_partss r, _ts_writez_clip_simple.._ts_writeJs *****r+*Indeterminate cursor sequence at position ,   rCrGrH)rr3rr0rr0r>r)'rlenfindrmatchragrouprrDr parserZr/rr&r'r1r(r2r4r5r6rfinditerrr from_params display_width make_sequence_text_sizing_clipextendr ValueErrornextr r rerSrh)!rBrCrDrQrGrHrErFrJr|outputrNidxr;r= current_stylecharnext_escmhl_statersgr_mrrrtsts_widthrseqnext_tabgrapheme grapheme_wrs! @r, _clip_simplers*h &FF C C*.N%)I*7A&&TM D //Cy #::$n44) yyq11H2~~C  6>>"(s33A  d###q ww|$$0]0}G` 1- K K !-#))MM!'')),,, -Ieegg+0;;;x &%h !#quuwwC  8/888MM!'')),,,%%''CCX!1!777+CCX!1!9991=(C+CCMM!*---MM!/222MM!+...$7)?)6%6)B%1%:1?%K%K66E,=mU[[]],[,[M(5II1=(C+C779---:''),,2''),,***$0 VVVW&&++O<<%<E#9#9*,H+++++++CX$!##CMM(+++$7)?)68OCeegg''))C 7=cBB   MM#   %%''C  4<<{{'S7]";<Hnn))))c))))) c***(;^-C-:N1HC Hnn ########## d### 1HC t3777888_(4666  ??!!!!c!!!!! h''' E\\cJ.#55 MM( # # # /!7!. 3YY3+e33 MM(c#sZ/?&@&@3uc??&RS T T T /!7!. z s8}}e D //h 776??NI 55r+rr write_cellsCallable[[str, int, int], None]c |}||kr-||z|kr$||||zS||ks ||z|kr||zStd||z } t|||z|z } jjg} jjdkrttj jjD]} | | ftjjt| z D]} | dfnAtj D],} | | t | ||zf-gd fd }|| z}d}| D]\}}||z}|| kr|}|| krnt|| t|| z }||kr3|dkr-s|t|| z} ||fn@|||t|| z}t|D]}|d ||z|}||||zS) z Emit tokens for a text-sizing (OSC 66) sequence, clipped to (start, end). Calls *write_cells(text, width, col)* for each emitted cell or sequence. Returns new column position. rrr flush_colr0r>rcsdSdD}tdD}tjjdkrt |ndjjjjjjjj}t|d |j  || dS)Ncg|] }|d S)rr*.0us r, z4_text_sizing_clip..flush..s---!1---r+c3&K|] }|dV dS)r Nr*rs r, z3_text_sizing_clip..flush..s&22qad222222r+rr)sumrrLr r numerator denominatorvertical_alignhorizontal_alignrrh terminatorrclear)rtextstotal_wrL pending_unitsscalerrs r,flushz _text_sizing_clip..flushs  F--}---22M22222! )/A--CJJJ1 I !6 I $bi&@ BB  vrwwu~~r} = = K K M M      r+r )rr0r>r) rrrSrerLrr rr rBraranger)rrNrCrDrErGrHrr rel_startrel_endunitsgrtr flush_col_posunit_pos unit_textunit_wunit_endoverlap abs_startirrs` ` @@r,rrs"00H e||h#-- B$$&&#666X~ czzS8^u,,X~Aus{##I#sX~&&,G IOE#%E yrw//AA % %A LL!U $ $ $ $ryU344 & &A LL"e % % % % & (( ? ?A LLE!_'35557<=> ? ? ? ?,.M )OMH" 6f$ y H  w   Eh((3x+C+CC f  !  ? #c(I&>&> >  )V!4 5 5 5 5 E- c(I666I7^^ 8 8 HaQ7777 E- >r+c ()*+,-|dk} i)t+g-d,d} d} d(d} rtnd*d} d+d,()*+,-fd }| t|krQ|| }| |kr|dvr| d| cxkr| krnn~t|d|d|d| krd} n-t j|| }|dn|} | dkr((n|d| dz}|dkrn|} |dkrtj || }|s$- | ,|f,dz ,| dz } &| durs*qt*| *(7| |kr1- | ,| f,dz ,*} |} tj| x}r~t!||||||||||  }|jt$jurD- | ,| f,dz ,|} n|jt$jur|j} n|jt$jur| |jz } |j} n- | ,|jf,dz ,||j|j| d| |jz } - | ,|jf,dz ,rB*@t9j|jD]&}t*| **} '|j} |j} W| dx}| dpd}| d}|Jt?tAj!||||}tE|| |||||} |} | }| r)tGj |rtId| d|| d x},|rtK|dz nd} |} q| d!x}x|rtK|nd}| |z}| |krB||krrcz|dkr'|r# | |f dz rdSt|D]}||z}|dkrE|dz dddkr" df|dz < |dz |dddkr" df|dz< |dz|d |||f|<|r*t|D]} ||z rdSdSdS)zIWrite *w* cells of text *s* at *write_col*, handling wide-char splitting.rr N)rr)rarrfdiscardpopadd)rrrroffsetsrc_colr;r[rrEhyperlink_cellsrQ seq_orderr]s r, _write_cellsz#_clip_painter.._write_cellss 66 3  )Y!:;;;Q  3^%;%2N FAhh - -F&(G{{uyy1g>>qAQFF&.]gk"''! 444yy'**1-22&.]gk"''! 444 IIgt $ $ $  # #G , , , ,q6i  8(( 8 8##I$67777  +^3*NNN + +33r+r} r~rPr rrT)rrrrrrrrhpa_n cforward_n cbackward_nz!Cursor left movement at position z would move z cells left from column z, exceeding string startrrrr)F) rr3rr0rr0rryr>r),setrrrSrfindrsearchrDrrrrarrr rrZr/rr&r'r1r(r2r4r5r7r6rrr8rrrrrrr0rrerr r rx).rBrCrDrQrGrHrErFrJr|rNrr= next_movementrrfoundrrrrrrrrrrrr n_forwardmove_endrr n_backwardrfill_colrr clip_startrr;r[rrrr]s. ` ` @@@@@@r, _clip_painterrs .h &F(*E #O,.II C C+/N&*I*7A&&TM#'M+0++++++++++++++: D //Cy #::$n44$](@(@(@(@S(@(@(@(@(@tzz&))4::d+;+;TZZ=O=OPPSVVV$&MM7>tSIIE*/-BBUYY[[Mq  !-99VS1W55r>> 6>>"(s33A   #y$!7888Q q ww|$$0]0}G` 1- K K !-#))$$c9aggii%@AAANI -Ieegg+0;;;x &%h !#quuwwC  8/888$$c9aggii%@AAANI%%''CCX!1!777+CCX!1!9991=(C+CC$$c9aj%ABBBNI L!/3.244441?*C$$c9ak%BCCCNI%6)B%1%:1?%K%K66E,=mU[[]],[,[M(5II,C+C779---:''),,2''),,***$0 VVVW&&(UC?  ""eegg''))C 7=cBB   )))6(-4c%jj1nn1eegg ggl333 @/9@C OOOq ?99E!1!1"3sE??C#4F4FGG55$ Xq!4444eegg !ww}555 B1<CS---! j3..$2C22%22?B222 z!77Ceegg   c9aggii8 9 9 9 NI%%''C  4<<C 1HC  6>>Qwwq 1HC  4<<{{'S7]";< %c#uoos8S7I7I J J33H La2222##########  #y$!7888Q 1HC t3777888_(4666  ??!!!!c!!!!!  #y(!;<<<Q E\\cJ.#55 L:s 3 3 3 3 3YY3+e33UCJCz)9 : :Z GHH ? ? Xq*v*=>>>> z s8}}q D //t ! 5#x H H I ''r+rPrTrF)rErFrGrQrJ overtypingrHrOptional[bool]c t|d}|dkr$|dkrtd|tj}||krdS|r|r |||St S|dkrM|durI|dkrCt|dkr0|rt ||||||||| } | | St|}d |v} |r| sd }|2|d ko*d |vp&d|vp"| o ttj |}n |r|d krd }|rtt|||||| ||| Stt|||||| ||| S)u Clip text to display columns (start, end) while preserving all terminal sequences. This function extracts a substring using visible column positions. Terminal escape sequences are preserved in output. If a wide character (width of 2) is split at a boundary, it is replaced with ``fillchar``. TAB characters (``\t``) are expanded to spaces up to the next tab stop, controlled by the ``tabsize`` parameter. When cursor movement is detected, a "painter's algorithm" is used unless ``overtyping=False`` is set. Cursor movement control codes are parsed for their effects instead of ignored. For these operations, it is assumed that ``text`` begins at column 0. For text containing **OSC 8 hyperlinks**, the visible text inside a hyperlink is clipped to the requested column range and the hyperlink sequence is rebuilt:: >>> clip('\x1b]8;;http://example.com\x07Click This link\x1b]8;;\x07', 6, 10) '\x1b]8;;http://example.com\x07This\x1b]8;;\x07' :param text: String to clip, may contain terminal escape sequences. :param start: Absolute starting column (inclusive, 0-indexed), default ``0``. :param end: Absolute ending column (exclusive). The default value, ``-1``, signifies "to the end of the line", clipping only from *start* without requiring the caller to measure the display width of *text*. :param fillchar: Character to use when a wide character must be split at a boundary (default space). Must have display width of 1. :param tabsize: Tab stop width (default 8). Set to 0 to pass tabs through as zero-width (preserved in output but don't advance column position). :param ambiguous_width: Width to use for East Asian Ambiguous (A) characters. Default is ``1`` (narrow). Set to ``2`` for CJK contexts. :param propagate_sgr: If True (default), SGR (terminal styling) sequences are propagated, matching :func:`propagate_sgr`. The result begins with any style active at the start position, retains any style changes occurring within the clipped region at their original position, and ends with a reset sequence if styles are active at the end position. :param control_codes: How to handle control characters and sequences: - ``'parse'`` (default): Track horizontal cursor movement and clip hyperlink text. Cursor overwrite is always allowed, with best effort results; indeterminate sequences (home, clear, reset, etc.) are preserved as zero-width. - ``'strict'``: Like ``parse``, but raises :exc:`ValueError` on sequences with indeterminate effects (cursor home, clear screen, reset, vertical movement, etc.) matching :func:`width` behavior. Also raises on out-of-bounds horizontal cursor movement. - ``'ignore'``: All control characters are treated as zero-width. Cursor movement is not tracked (fastest path). :param overtyping: Whether to use the painter's algorithm for cursor movement (``\b`` backspace, ``\r`` carriage return, and CSI cursor left/right/position sequences). When ``None`` (default), auto-detects by scanning for these characters in *text*. Set to ``False`` for improved performance when the caller knows *text* contains no cursor movement characters. Set to ``True`` to force the painter's algorithm (useful for testing). Has no effect when ``control_codes='ignore'``. :param term_program: Terminal software identifier for table correction. ``False`` (default) disables override lookup. ``True`` reads the ``TERM_PROGRAM`` or ``TERM`` environment variable for auto-detection. Accepts a canonical terminal name matching :func:`list_term_programs`, such as from XTVERSION_, ENQ_, or ``TERM_PROGRAM``. .. versionadded:: 0.8.0 :returns: Substring of ``text`` spanning display columns (start, end), with all terminal sequences preserved and wide characters at boundaries replaced with ``fillchar``. :raises ValueError: If ``end`` is negative and not ``-1``, or if ``control_codes='strict'`` and an indeterminate-effect sequence or out-of-bounds cursor movement is encountered. SGR (terminal styling) sequences are propagated by default. The result begins with any active style and ends with a reset:: >>> clip('\x1b[1;34mHello world\x1b[0m', 6, 11) '\x1b[1;34mworld\x1b[0m' >>> clip('\x1b[1mbold\x1b[m normal', 1, 9) '\x1b[1mold\x1b[m norm' Set ``propagate_sgr=False`` to disable this behavior. .. note:: Segmentation follows UAX #29 grapheme clusters. Tamil, Kannada and Sinhala may be clipped short of ``end`` and leave a bare virama at the boundary. Correct rendering of these languages is not specified by unicode.org standards, and no terminal emulator or editor handles them legibly, so this library function makes no attempt at accommodating them. .. versionadded:: 0.3.0 .. versionchanged:: 0.5.0 Added ``propagate_sgr`` parameter (default True). .. versionchanged:: 0.7.0 Added ``control_codes`` parameter (default 'parse'). OSC 8 hyperlink-aware clipping. OSC 66 text sizing protocol support. Added ``overtyping`` parameter (default None, auto-detect). .. versionchanged:: 0.8.4 ``start`` now defaults to ``0`` and ``end`` to ``-1``, meaning "to the end of the line":: >>> clip('\x1b[1;34mHello world\x1b[0m', 6) '\x1b[1;34mworld\x1b[0m' Example:: >>> clip('hello world', 0, 5) 'hello' >>> clip('中文字', 0, 3) # Wide char split at column 3 '中 ' >>> clip('a\tb', 0, 10) # Tab expanded to spaces 'a b' rrPz5end must be -1 (to end of line) or non-negative, got rNrTr )rErFrGrQrJrHr~Fignorerr) rBrCrDrQrGrHrErFrJ)rSrsysmaxsizeisascii isprintablerrrryrrrArr) rBrCrDrErFrGrQrJrrHr:has_escs r,rrs%| qMME Qww "99MMMOO Ok e||r ||~~$**,,E#I (($&&! H ""x'7'7'9'9" %x+='lDDD  M,_==OnGW  X % K t^Jtt|JH$:A$GGHH   11    '+%' ! ! !   L  #'!#     r+r)r:r3r;r<r=r<r>r3)rBr3rCr0rDr0rEr3rFr0rGr0rHrIrJrKrLr rMr0rNr0r>r.) r[r\r]r^rCr0rDr0rEr3r>r3)rBr3rCr0rDr0rQryrGr0rHrIrEr3rFr0rJrKr>rz)rrrNr0rCr0rDr0rEr3rGr0rHrIrrr>r0)rrP)rBr3rCr0rDr0rEr3rFr0rGr0rQryrJrKrrrHrIr>r3)5r# __future__rosrr$ itertoolsrtypingrrrr_widthr rr hyperlinkr r sgr_staterrrrrr _constantsr text_sizingrrescape_sequencesrrrr9environrfr _wcwidth_cr _c_clip_impl ImportErrorEnumrr.rArZrxrrrr*r+r,r s000"""""" ::::::::::::$$$$$$111111110000000000000000 /.....55555555>>>>>>>>>>0///:>>"B'' GG444444  tyz$6:,CCCCL4444nr6r6r6r6jUUUUpG'G'G'G'XC :A!%$CCCCCCCCsBBB