i'RdZddlZddlmZmZmZmZddlmZddddddd d d Z d e d e fdZ d1de de de de d ee e ee ee ff dZde deee e fde d e fdZde de d eee e ffdZde de d eee e ffdZde de d eee e ffdZde de d eee e ffdZde de d eee e ffdZde de d eee e ffdZde d ee fdZd ee d!eee e fd eee e ffd"Zde de d eee e ffd#Zde de d eee e ffd$Zde de d eee e ffd%Zd&ee d'e d(e d)e d ee e ff d*Zde d&ee d+ee de d,e d eee e ff d-Zde d.e d/eee e fd eee e ffd0ZdS)2aP Fuzzy Matching Module for File Operations Implements a multi-strategy matching chain to robustly find and replace text, accommodating variations in whitespace, indentation, and escaping common in LLM-generated code. The 8-strategy chain (inspired by OpenCode), tried in order: 1. Exact match - Direct string comparison 2. Line-trimmed - Strip leading/trailing whitespace per line 3. Whitespace normalized - Collapse multiple spaces/tabs to single space 4. Indentation flexible - Ignore indentation differences entirely 5. Escape normalized - Convert \n literals to actual newlines 6. Trimmed boundary - Trim first/last line whitespace only 7. Block anchor - Match first+last lines, use similarity for middle 8. Context-aware - 50% line similarity threshold Multi-occurrence matching is handled via the replace_all flag. Usage: from tools.fuzzy_match import fuzzy_find_and_replace new_content, match_count, strategy, error = fuzzy_find_and_replace( content="def foo():\n pass", old_string="def foo():", new_string="def bar():", replace_all=False ) N)TupleOptionalListCallable)SequenceMatcher"'z---z... )u“u”u‘u’u—u–u… textreturncptD]\}}|||}|S)zBNormalizes Unicode characters to their standard ASCII equivalents.) UNICODE_MAPitemsreplace)r charrepls 9/home/agentuser/.hermes/hermes-agent/tools/fuzzy_match.py_unicode_normalizer+s;!''))(( d||D$'' KFcontent old_string new_string replace_allc |s|dddfS||kr|dddfSdtfdtfdtfdtfd tfd t fd t fd tfd tfg }|D]g\}}|||}|rTt|dkr|s|dddt|dfcSt|||}|t||dfcSh|dddfS)a) Find and replace text using a chain of increasingly fuzzy matching strategies. Args: content: The file content to search in old_string: The text to find new_string: The replacement text replace_all: If True, replace all occurrences; if False, require uniqueness Returns: Tuple of (new_content, match_count, strategy_name, error_message) - If successful: (modified_content, number_of_replacements, strategy_used, None) - If failed: (original_content, 0, None, error_description) rNzold_string cannot be emptyz'old_string and new_string are identicalexact line_trimmedwhitespace_normalizedindentation_flexibleescape_normalizedtrimmed_boundaryunicode_normalized block_anchor context_awarezFound zY matches for old_string. Provide more context to make it unique, or use replace_all=True.z1Could not find a match for old_string in the file) _strategy_exact_strategy_line_trimmed_strategy_whitespace_normalized_strategy_indentation_flexible_strategy_escape_normalized_strategy_trimmed_boundary_strategy_unicode_normalized_strategy_block_anchor_strategy_context_awarelen_apply_replacements) rrrr strategies strategy_name strategy_fnmatches new_contents rfuzzy_find_and_replacer72sY >4!===Z4!JJJ /" /0 "AB !?@ 9: 78 ;< /0 12 .J'1 B B" {+gz22  B7||a 4XS\\XXX .gw KKKG mTA A A A B AtP PPrr5cpt|dd}|}|D]\}}|d||z||dz}|S)a Apply replacements at the given positions. Args: content: Original content matches: List of (start, end) positions to replace new_string: Replacement text Returns: Content with replacements applied c|dS)Nr)xs rz%_apply_replacements..vs 1Q4rT)keyreverseN)sorted)rr5rsorted_matchesresultstartends rr1r1hsXGFFFN F$<< s*,vcdd|; Mrpatterncg}d} |||}|dkrn-|||t|zf|dz}J|S)zStrategy 1: Exact string match.rTr&)findappendr0)rrDr5rBposs rr'r'sgG Ell7E** "99 S3w<</0111a  Nrcd|dD}d|}|d}d|D}t|||||S)z Strategy 2: Match with line-by-line whitespace trimming. Strips leading/trailing whitespace from each line before matching. c6g|]}|Sr:strip.0lines r z*_strategy_line_trimmed..s BBBdTZZ\\BBBr c6g|]}|Sr:rLrNs rrQz*_strategy_line_trimmed..s GGG GGGr)splitjoin_find_normalized_matches)rrD pattern_linespattern_normalized content_linescontent_normalized_liness rr(r(s}CBgmmD.A.ABBBM=11MM$''MGGGGG $ 8#  rc~d}||}||}t||}|sgSt|||S)zC Strategy 3: Collapse multiple whitespace to single space. c.tjdd|S)Nz[ \t]+r )resubss r normalizez2_strategy_whitespace_normalized..normalizesvia(((r)r'_map_normalized_positions)rrDrarXcontent_normalizedmatches_in_normalizeds rr)r)sj)))#7++"7++,,>@RSS   %W.@BW X XXrc |d}d|D}d|dD}t||||d|S)z Strategy 4: Ignore indentation differences entirely. Strips all leading whitespace from lines before matching. rRc6g|]}|Sr:lstriprNs rrQz2_strategy_indentation_flexible..s FFFdkkmmFFFrc6g|]}|Sr:rgrNs rrQz2_strategy_indentation_flexible..s CCCtT[[]]CCCr)rTrVrU)rrDrYcontent_stripped_linesrWs rr*r*ss MM$''MFF FFFCCw}}T/B/BCCCM # 6=))  rcNd}||}||krgSt||S)zt Strategy 5: Convert escape sequences to actual characters. Handles \n -> newline, \t -> tab, etc. c~|ddddddS)Nz\nrRz\t z\r )rr_s runescapez-_strategy_escape_normalized..unescapes6yy%%--eT::BB5$OOOr)r')rrDropattern_unescapeds rr+r+sG PPP!))G## 7$5 6 66rc |d}|sgS|d|d<t|dkr|d|d<d|}|d}g}t|}t t||z dzD]}||||z}|} | d| d<t| dkr| d| d<d| |kr ? ?11#Aa*<&<$<= "&&(( $Q--// A {  a  )"o3355KO 99[ ! !%5 5 5!:q!&8"8#g,,"" Iw NNIw/ 0 0 0 Nroriginalcg}d}|D]G}||t|}||t|ndz }H|||S)uBuild a list mapping each original character index to its normalized index. Because UNICODE_MAP replacements may expand characters (e.g. em-dash → '--', ellipsis → '...'), the normalised string can be longer than the original. This map lets us convert positions in the normalised string back to the corresponding positions in the original string. Returns a list of length ``len(original) + 1``; entry ``i`` is the normalised index that character ``i`` maps to. rNr&)rHrgetr0)r|rAnorm_posrrs r_build_orig_to_norm_maprsrFH99 ht$$!1CIIIq8 MM( Mr orig_to_norm norm_matchesc,i}t|ddD]\}}||vr|||<g}t|dz }|D]T\}}||vr ||} | } | |kr#|| |kr| dz } | |kr || |k|| | fU|S)zNConvert (start, end) positions in the normalised string to original positions.NrFr&) enumerater0rH) rrnorm_to_orig_startorig_posrresultsorig_len norm_startnorm_end orig_startorig_ends r_map_positions_norm_to_origrs *,' SbS(9::44( - - -+3 x (%'G<  1$H , / / H / / / ' 3 !!l8&>L J-lLII  *733L &|\ B BBrc t|}t|}|d}t|dkrgS|d}|d}|d}|d}t|} g} t t|| z dzD]Y} || |kr9|| | zdz |kr| | Zg} t| } | dkrdnd}| D]} | dkrd}nfd|| dz| | zdz }d|dd}td ||}||krRST>T:T'UVVN!YY}QrT':;;N(~~NNTTVVJ  " "!:"Aq+='=s7||"" Iw NNIw/ 0 0 0 Nrc `|d}|d}|sgSg}t|}tt||z dzD]}||||z}d}t||D]W\} } t d| | } | dkr|dz }X|t|dzkr ? ?11#Aa*<&<$<= !"!-== + +NFF!$  GGMMOOCd{{%*% !C $6$6$< < z,_calculate_line_positions..s,IIdCIIMIIIIIIrNc3:K|]}t|dzVdSrrrNs rrz,_calculate_line_positions..s,EED#d))a-EEEEEErr&)sum)rYrrrrzr{s rrtrtsnIImKZK.HIIIIIIEEM)8),DEEEEEIG.    g rrZrXc `|d}t|}g}tt||z dzD]d}d||||z} | |kr.(s+VV41aa:ooQooooVVr)r0rHrmin)r|rrrorig_idxnorm_idxrnorm_to_orig_endrroriginal_matchesrrrrs @rrbrbs   LHH S]] " "x#j//'A'A H H!5 5 5    ) ) ) MH MHH h 5 ( (Z-AS-H-H    ) ) ) MH#h--''HX,>e,K,KA  h 5 ( (    ) ) ) MHH    ) ) ) MH' S]] " "x#j//'A'A, S]] " "C OO,,,A  S]] " " ' 55..( - - -+3 x (%-"" 2LL H + + ++J7JJVVVV9\+B+BVVVVVJ a<+ + +'1 59HH!X %:;HX&&8H+=+F+F MHX&&8H+=+F+F S3x==-I-I JKKKK r)F)__doc__r]typingrrrrdifflibrrstrrboolintr7r1r'r(r)r*r+r,rrr-r.r/rtrVrbr:rrrs< 222222222222######SScs  SS163Q3QC3QS3Qc3Q)-3Q:?S(SV-YabeYf@f:g3Q3Q3Q3QltE#s(O/DRUZ]6 S 3 4c3h3H    C#$uS#X:O(YSY3Y4cSVhCXYYYY* C # $uSRUXBW     77s7tE#s(O?T7777&''c'd5c?>S''''Tcd3i*s)uS#X' %S/:C#CCU3PS8_@UCCCC>5C5#5$uS#X:O5555p S 3 4c3h;P    NT#YC(+=@EJ3PS8_( c $s) 8