Module: DataRedactor

Defined in:
lib/data_redactor.rb,
lib/data_redactor/version.rb,
ext/data_redactor/data_redactor.c

Defined Under Namespace

Classes: InvalidPatternError, UnknownTagError

Constant Summary collapse

TAGS = 
{
  credentials: TAG_CREDENTIALS,
  financial:   TAG_FINANCIAL,
  tax_id:      TAG_TAX_ID,
  national_id: TAG_NATIONAL_ID,
  contact:     TAG_CONTACT,
  network:     TAG_NETWORK,
  travel:      TAG_TRAVEL,
  other:       TAG_OTHER,
  custom:      TAG_CUSTOM
}.freeze
CAPTURE_GROUP_RE =

Capture groups break boundary-wrapper group index assumptions ([1],,[3] shift).

/(?<!\\)\((?!\?:)/.freeze
RUBY_ONLY_SYNTAX_RE =

Ruby regex syntax that has no POSIX ERE equivalent.

/\\[dDwWsShHbB]|\(\?[<!=]|\(\?<[a-zA-Z]|\(\?[imx]|[*+?]\?/.freeze
PLACEHOLDER_DEFAULT = 
"[REDACTED]"
VERSION = 
"0.5.0"
PH_MODE_PLAIN =

Placeholder mode constants.

INT2NUM(PLACEHOLDER_MODE_PLAIN)
PH_MODE_TAGGED = 
INT2NUM(PLACEHOLDER_MODE_TAGGED)
PH_MODE_HASH = 
INT2NUM(PLACEHOLDER_MODE_HASH)
TAG_CREDENTIALS =

Expose tag bitmask values so the Ruby wrapper can build the mask.

INT2NUM(TAG_CREDENTIALS)
TAG_FINANCIAL = 
INT2NUM(TAG_FINANCIAL)
TAG_TAX_ID = 
INT2NUM(TAG_TAX_ID)
TAG_NATIONAL_ID = 
INT2NUM(TAG_NATIONAL_ID)
TAG_CONTACT = 
INT2NUM(TAG_CONTACT)
TAG_NETWORK = 
INT2NUM(TAG_NETWORK)
TAG_TRAVEL = 
INT2NUM(TAG_TRAVEL)
TAG_OTHER = 
INT2NUM(TAG_OTHER)
TAG_CUSTOM = 
INT2NUM(TAG_CUSTOM)
TAG_ALL = 
INT2NUM(TAG_ALL)

Class Method Summary collapse

Class Method Details

._add_pattern(rb_name, rb_source, rb_tag_bit, rb_boundary) ⇒ Object

DataRedactor._add_pattern(name, source, tag_bit, boundary) -> nil

Compile ‘source` as POSIX ERE (with boundary wrapper when boundary=1), store under `name`. Replaces any existing pattern with the same name. Raises DataRedactor::InvalidPatternError on regcomp failure.



550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
 
# File 'ext/data_redactor/data_redactor.c', line 550

static VALUE rb_add_pattern(VALUE self, VALUE rb_name, VALUE rb_source,
                             VALUE rb_tag_bit, VALUE rb_boundary) {
    Check_Type(rb_name,   T_STRING);
    Check_Type(rb_source, T_STRING);

    const char *name    = StringValueCStr(rb_name);
    const char *source  = StringValueCStr(rb_source);
    int tag_bit         = NUM2INT(rb_tag_bit);
    int boundary        = NUM2INT(rb_boundary);

    /* Build the pattern string (wrap boundary if requested) */
    char *pat_to_compile;
    char *wrapped = NULL;
    if (boundary) {
        wrapped = wrap_boundary(source);
        if (!wrapped) rb_raise(rb_eNoMemError, "wrap_boundary allocation failed");
        pat_to_compile = wrapped;
    } else {
        pat_to_compile = (char *)source;
    }

    regex_t compiled;
    int ret = regcomp(&compiled, pat_to_compile, REG_EXTENDED);
    free(wrapped);

    if (ret != 0) {
        char errbuf[256];
        regerror(ret, &compiled, errbuf, sizeof(errbuf));
        regfree(&compiled);
        VALUE eClass = rb_const_get(rb_define_module("DataRedactor"),
                                    rb_intern("InvalidPatternError"));
        rb_raise(eClass, "%s", errbuf);
    }

    /* Replace existing or append */
    int idx = find_custom_by_name(name);
    if (idx >= 0) {
        free_custom_at(idx);
    } else {
        if (custom_count >= custom_cap) {
            int new_cap = custom_cap == 0 ? 8 : custom_cap * 2;
            custom_pattern_t *tmp = (custom_pattern_t *)realloc(
                custom_patterns, sizeof(custom_pattern_t) * new_cap);
            if (!tmp) {
                regfree(&compiled);
                rb_raise(rb_eNoMemError, "custom_patterns realloc failed");
            }
            custom_patterns = tmp;
            custom_cap = new_cap;
        }
        idx = custom_count++;
    }

    custom_patterns[idx].name     = strdup(name);
    custom_patterns[idx].source   = strdup(source);
    custom_patterns[idx].compiled = compiled;
    custom_patterns[idx].tag      = tag_bit;
    custom_patterns[idx].boundary = boundary;

    if (!custom_patterns[idx].name || !custom_patterns[idx].source) {
        rb_raise(rb_eNoMemError, "strdup failed");
    }

    return Qnil;
}

._clear_custom_patterns ⇒ Object

DataRedactor._clear_custom_patterns -> nil



642
643
644
645
646
647
648
 
# File 'ext/data_redactor/data_redactor.c', line 642

static VALUE rb_clear_custom_patterns(VALUE self) {
    for (int i = 0; i < custom_count; i++) {
        free_custom_at(i);
    }
    custom_count = 0;
    return Qnil;
}

._custom_patterns ⇒ Object

DataRedactor._custom_patterns -> Array<Hash>

Returns [source:, tag_bit:, boundary:, …] for each custom pattern.



655
656
657
658
659
660
661
662
663
664
665
666
 
# File 'ext/data_redactor/data_redactor.c', line 655

static VALUE rb_custom_patterns(VALUE self) {
    VALUE arr = rb_ary_new_capa(custom_count);
    for (int i = 0; i < custom_count; i++) {
        VALUE h = rb_hash_new();
        rb_hash_aset(h, ID2SYM(rb_intern("name")),     rb_str_new_cstr(custom_patterns[i].name));
        rb_hash_aset(h, ID2SYM(rb_intern("source")),   rb_str_new_cstr(custom_patterns[i].source));
        rb_hash_aset(h, ID2SYM(rb_intern("tag_bit")),  INT2NUM(custom_patterns[i].tag));
        rb_hash_aset(h, ID2SYM(rb_intern("boundary")), custom_patterns[i].boundary ? Qtrue : Qfalse);
        rb_ary_push(arr, h);
    }
    return arr;
}

._redact(rb_text, rb_mask, rb_ph_mode, rb_ph_str) ⇒ Object

DataRedactor._redact(text, mask, ph_mode, ph_str) -> String

‘mask` — integer bitmask of TAG_* values (only / except filtering). `ph_mode` — 0 = plain string, 1 = tagged “[REDACTED:TAG]”, 2 = hash “[TAG_xxxx]”. `ph_str` — the plain string for mode 0; ignored for modes 1 and 2.

The Ruby wrapper builds all four arguments and is the public API.



796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
 
# File 'ext/data_redactor/data_redactor.c', line 796

static VALUE rb_data_redactor_redact(VALUE self, VALUE rb_text, VALUE rb_mask,
                                     VALUE rb_ph_mode, VALUE rb_ph_str) {
    Check_Type(rb_text,   T_STRING);
    Check_Type(rb_ph_str, T_STRING);

    int mask    = NUM2INT(rb_mask);
    int ph_mode = NUM2INT(rb_ph_mode);
    const char *ph_str_plain = StringValueCStr(rb_ph_str);

    const char *input = StringValueCStr(rb_text);
    char *working = strdup(input);
    if (!working) rb_raise(rb_eNoMemError, "strdup failed");

    placeholder_t ph;
    ph.mode = ph_mode;

    for (int i = 0; i < NUM_PATTERNS; i++) {
        if ((pattern_tags[i] & mask) == 0) continue;
        ph.str = (ph_mode == PLACEHOLDER_MODE_PLAIN)
                     ? ph_str_plain
                     : tag_name_for_bit(pattern_tags[i]);
        char *result = replace_all_matches(&compiled_patterns[i], working,
                                           boundary_wrapped[i], &ph);
        free(working);
        if (!result) rb_raise(rb_eNoMemError, "replace_all_matches allocation failed");
        working = result;
    }

    for (int i = 0; i < custom_count; i++) {
        if ((custom_patterns[i].tag & mask) == 0) continue;
        ph.str = (ph_mode == PLACEHOLDER_MODE_PLAIN)
                     ? ph_str_plain
                     : tag_name_for_bit(custom_patterns[i].tag);
        char *result = replace_all_matches(&custom_patterns[i].compiled, working,
                                           custom_patterns[i].boundary, &ph);
        free(working);
        if (!result) rb_raise(rb_eNoMemError, "replace_all_matches allocation failed (custom)");
        working = result;
    }

    VALUE rb_result = rb_str_new_cstr(working);
    free(working);
    return rb_result;
}

._remove_pattern(rb_name) ⇒ Object

DataRedactor._remove_pattern(name) -> true/false

Remove the named custom pattern. Returns true if found and removed.



621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
 
# File 'ext/data_redactor/data_redactor.c', line 621

static VALUE rb_remove_pattern(VALUE self, VALUE rb_name) {
    Check_Type(rb_name, T_STRING);
    const char *name = StringValueCStr(rb_name);

    int idx = find_custom_by_name(name);
    if (idx < 0) return Qfalse;

    free_custom_at(idx);

    /* Shift remaining entries left */
    for (int i = idx; i < custom_count - 1; i++) {
        custom_patterns[i] = custom_patterns[i + 1];
    }
    custom_count--;

    return Qtrue;
}

._scan(rb_text, rb_mask) ⇒ Object

DataRedactor._scan(text, mask) -> Hash

Returns { redacted: String, matches: Array<Hash> } where each match hash is:

{ tag: Symbol, name: String, value: String, start: Integer, length: Integer }

Matches are reported in the order they are consumed by the sequential redaction loop (built-ins first, most-specific to most-generic; then custom patterns). ‘start` and `length` refer to byte positions in the original input string. Because patterns run sequentially on a shrinking/expanding working buffer, positions are tracked relative to the original by maintaining a running offset.



853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
 
# File 'ext/data_redactor/data_redactor.c', line 853

static VALUE rb_data_redactor_scan(VALUE self, VALUE rb_text, VALUE rb_mask) {
    Check_Type(rb_text, T_STRING);
    int mask = NUM2INT(rb_mask);

    const char *input     = StringValueCStr(rb_text);
    size_t      input_len = strlen(input);

    /* Working buffer — we redact with the default plain placeholder so the
     * scan result also contains the redacted string.                        */
    static const placeholder_t ph_default = { PLACEHOLDER_MODE_PLAIN, "[REDACTED]" };

    char *working = strdup(input);
    if (!working) rb_raise(rb_eNoMemError, "strdup failed");

    VALUE matches_arr = rb_ary_new();

    /*
     * To map working-buffer positions back to original-string positions we
     * maintain a log of every replacement already applied.  Each entry records
     * where in the *working* buffer the replacement started (after all prior
     * replacements) and how many bytes were removed (orig_len) vs. inserted
     * (always 10, the length of "[REDACTED]").
     *
     * For a new match at working position W:
     *   cumulative_shift_before_W = sum of (10 - orig_len) for all prior
     *                               replacements whose working_pos <= W
     *   original_pos = W - cumulative_shift_before_W
     *
     * Replacements are appended in order so the log is already sorted by
     * working_pos; we just walk it linearly per match.
     */
    typedef struct { long wpos; long orig_len; } repl_t;
    repl_t *repl_log = NULL;
    int     repl_count = 0;
    int     repl_cap   = 0;

    #define REPL_LOG_PUSH(_wpos, _olen) do {                                  \
        if (repl_count >= repl_cap) {                                         \
            int _nc = repl_cap == 0 ? 16 : repl_cap * 2;                     \
            repl_t *_t = (repl_t *)realloc(repl_log, sizeof(repl_t) * _nc);  \
            if (!_t) { free(repl_log); free(working); rb_raise(rb_eNoMemError, "repl_log"); } \
            repl_log = _t; repl_cap = _nc;                                    \
        }                                                                     \
        repl_log[repl_count].wpos     = (_wpos);                              \
        repl_log[repl_count].orig_len = (_olen);                              \
        repl_count++;                                                         \
    } while (0)

    /* Map a position in the current working buffer to original-string position. */
    #define WORKING_TO_ORIG(_wpos) ({                                         \
        long _shift = 0;                                                      \
        for (int _ri = 0; _ri < repl_count; _ri++) {                         \
            if (repl_log[_ri].wpos <= (_wpos))                                \
                _shift += 10 - repl_log[_ri].orig_len;                       \
        }                                                                     \
        (_wpos) - _shift;                                                     \
    })

    /* Collect matches for one pattern on the current working buffer, translate
     * positions to original coordinates, then do the replacement.            */
    #define COLLECT_AND_REPLACE(pat, use_bnd, tag_bit, pat_name) do {        \
        const char *_cur = working;                                           \
        regmatch_t _m[4];                                                     \
        while (regexec((pat), _cur, 4, _m, 0) == 0) {                        \
            regoff_t _fso = _m[0].rm_so, _feo = _m[0].rm_eo;                 \
            if (_fso < 0 || _feo < _fso) break;                               \
            regoff_t _cso = _fso, _ceo = _feo;                                \
            if (use_bnd) {                                                    \
                if (_m[1].rm_so >= 0 && _m[1].rm_eo > _m[1].rm_so)          \
                    _cso = _m[1].rm_eo;                                       \
                if (_m[3].rm_so >= 0 && _m[3].rm_eo > _m[3].rm_so)          \
                    _ceo = _m[3].rm_so;                                       \
            }                                                                 \
            size_t _vlen = (size_t)(_ceo - _cso);                             \
            long _wpos   = (long)(_cur - working) + (long)_cso;              \
            long _orig   = WORKING_TO_ORIG(_wpos);                            \
            VALUE _match = rb_hash_new();                                     \
            rb_hash_aset(_match, ID2SYM(rb_intern("tag")),                    \
                         ID2SYM(rb_intern(tag_name_for_bit(tag_bit))));       \
            rb_hash_aset(_match, ID2SYM(rb_intern("name")),                  \
                         rb_str_new_cstr(pat_name));                          \
            rb_hash_aset(_match, ID2SYM(rb_intern("value")),                 \
                         rb_str_new(_cur + _cso, _vlen));                     \
            rb_hash_aset(_match, ID2SYM(rb_intern("start")),                 \
                         LONG2NUM(_orig));                                    \
            rb_hash_aset(_match, ID2SYM(rb_intern("length")),                \
                         LONG2NUM((long)_vlen));                              \
            rb_ary_push(matches_arr, _match);                                 \
            /* Log this replacement; wpos advances by 10 for subsequent entries */ \
            REPL_LOG_PUSH(_wpos, (long)_vlen);                                \
            /* Re-anchor cursor: skip past the full match in working buf */   \
            if (_feo == _fso) { if (*_cur) _cur++; else break; }             \
            else _cur += _feo;                                                \
        }                                                                     \
        char *_next = replace_all_matches((pat), working, (use_bnd), &ph_default); \
        free(working);                                                        \
        if (!_next) { free(repl_log); rb_raise(rb_eNoMemError, "replace_all_matches failed in scan"); } \
        working = _next;                                                      \
    } while (0)

    for (int i = 0; i < NUM_PATTERNS; i++) {
        if ((pattern_tags[i] & mask) == 0) continue;
        COLLECT_AND_REPLACE(&compiled_patterns[i], boundary_wrapped[i],
                            pattern_tags[i], pattern_names[i]);
    }

    for (int i = 0; i < custom_count; i++) {
        if ((custom_patterns[i].tag & mask) == 0) continue;
        COLLECT_AND_REPLACE(&custom_patterns[i].compiled,
                            custom_patterns[i].boundary,
                            custom_patterns[i].tag, custom_patterns[i].name);
    }

    #undef COLLECT_AND_REPLACE
    #undef WORKING_TO_ORIG
    #undef REPL_LOG_PUSH

    free(repl_log);

    VALUE result = rb_hash_new();
    VALUE rb_redacted = rb_str_new_cstr(working);
    free(working);
    rb_hash_aset(result, ID2SYM(rb_intern("redacted")), rb_redacted);
    rb_hash_aset(result, ID2SYM(rb_intern("matches")),  matches_arr);
    return result;

    (void)input_len; /* suppress unused-variable warning */
}

.add_pattern(name:, regex:, tag: :custom, boundary: false) ⇒ Object

Add (or replace) a custom redaction pattern.

name: unique identifier string regex: String (POSIX ERE) or Regexp; Ruby-only syntax raises InvalidPatternError tag: one of the TAGS keys (default :custom), or any built-in tag boundary: wrap with word-boundary guards; incompatible with capture groups

Raises:

  • (ArgumentError) 


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
 
# File 'lib/data_redactor.rb', line 81

def add_pattern(name:, regex:, tag: :custom, boundary: false)
  raise ArgumentError, "name must be a non-empty String" \
    unless name.is_a?(String) && !name.empty?

  source = case regex
           when String then regex
           when Regexp then regex.source
           else raise ArgumentError, "regex must be a String or Regexp, got #{regex.class}"
           end

  if source =~ RUBY_ONLY_SYNTAX_RE
    raise InvalidPatternError,
      "pattern #{name.inspect} uses Ruby-only syntax (#{$&.inspect}); " \
      "use POSIX ERE — no \\d, \\s, \\w, \\b, lookaround, non-greedy, or named groups"
  end

  if boundary && source =~ CAPTURE_GROUP_RE
    raise InvalidPatternError,
      "pattern #{name.inspect} has capture groups and cannot use boundary: true"
  end

  tag_bit = TAGS[tag] or raise UnknownTagError,
    "unknown tag #{tag.inspect}; valid tags: #{TAGS.keys.inspect}"

  _add_pattern(name, source, tag_bit, boundary ? 1 : 0)
end

.bits_for(tag_list) ⇒ Object 



123
124
125
126
127
128
129
 
# File 'lib/data_redactor.rb', line 123

def bits_for(tag_list)
  Array(tag_list).inject(0) do |acc, tag|
    bit = TAGS[tag] or raise UnknownTagError,
      "unknown tag #{tag.inspect}; valid tags: #{TAGS.keys.inspect}"
    acc | bit
  end
end

.clear_custom_patterns! ⇒ Object 



119
120
121
 
# File 'lib/data_redactor.rb', line 119

def clear_custom_patterns!
  _clear_custom_patterns
end

.custom_patterns ⇒ Object 



112
113
114
115
116
117
 
# File 'lib/data_redactor.rb', line 112

def custom_patterns
  _custom_patterns.map do |h|
    { name: h[:name], source: h[:source], tag: TAGS.key(h[:tag_bit]) || :custom,
      boundary: h[:boundary] }
  end
end

.redact(text, only: nil, except: nil, placeholder: PLACEHOLDER_DEFAULT) ⇒ Object

Raises:

  • (ArgumentError) 


34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
 
# File 'lib/data_redactor.rb', line 34

def redact(text, only: nil, except: nil, placeholder: PLACEHOLDER_DEFAULT)
  raise ArgumentError, "pass only: or except:, not both" if only && except

  mask =
    if only
      bits_for(only)
    elsif except
      TAG_ALL & ~bits_for(except)
    else
      TAG_ALL
    end

  ph_mode, ph_str = resolve_placeholder(placeholder)
  _redact(text, mask, ph_mode, ph_str)
end

.remove_pattern(name) ⇒ Object 



108
109
110
 
# File 'lib/data_redactor.rb', line 108

def remove_pattern(name)
  _remove_pattern(name.to_s)
end

.resolve_placeholder(placeholder) ⇒ Object

Returns [ph_mode_int, ph_str] for the C layer.

placeholder: "***"      -> plain string
placeholder: :tagged    -> "[REDACTED:TAGNAME]"
placeholder: :hash      -> "[TAGNAME_xxxx]"



135
136
137
138
139
140
141
142
143
144
 
# File 'lib/data_redactor.rb', line 135

def resolve_placeholder(placeholder)
  case placeholder
  when :tagged then [PH_MODE_TAGGED, ""]
  when :hash   then [PH_MODE_HASH,   ""]
  when String  then [PH_MODE_PLAIN,  placeholder]
  else
    raise ArgumentError,
      "placeholder must be a String, :tagged, or :hash — got #{placeholder.inspect}"
  end
end

.scan(text, only: nil, except: nil) ⇒ Object

Scan text without necessarily redacting it.

Returns { redacted: String, matches: [name:, value:, start:, length:, …] } The :tag value is a Symbol matching one of DataRedactor.tags. :start and :length are byte offsets into the original string.

Raises:

  • (ArgumentError) 


55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
 
# File 'lib/data_redactor.rb', line 55

def scan(text, only: nil, except: nil)
  raise ArgumentError, "pass only: or except:, not both" if only && except

  mask =
    if only
      bits_for(only)
    elsif except
      TAG_ALL & ~bits_for(except)
    else
      TAG_ALL
    end

  result = _scan(text, mask)
  # Normalise: convert tag string from C (uppercase) back to the Symbol used in TAGS
  result[:matches].each do |m|
    m[:tag] = m[:tag].to_s.downcase.to_sym
  end
  result
end

.tags ⇒ Object 



30
31
32
 
# File 'lib/data_redactor.rb', line 30

def tags
  TAGS.keys
end