1 contributor
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
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
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
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
182
183
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
#!/bin/bash
set -e
manuscrit=${1:?}
ancien=${2:?}
nouveau=${3:?}
if test "${manuscrit:0:1}" != '/'; then
manuscrit="$PWD/$manuscrit"
fi
test -r "$manuscrit" || exit 1
test $(file --brief --mime-type --dereference "$manuscrit") == 'application/zip' || exit 2
backup="${manuscrit/%.msk} (avant renommage de «${ancien}» en «${nouveau}»).msk"
function trap_exit () {
rm -fr $temp
cd - > /dev/null
}
function determinant () {
eval "local nom=\"\$$1\""
if [[ ${nom:0:1} == @(A|E|H|I|O|U) \
|| ${nom:0:1} == @(Â|Ê|H|Î|Ô|Û) \
|| ${nom:0:1} == @(Ä|Ë|H|Ï|Ö|Ü) \
|| ${nom:0:1} == @(À|È|H|Ì|Ò|Ù) \
|| ${nom:0:1} == @(Á|É|H|Í|Ó|Ú) \
]]; then
eval "determinant_$1=\"d'\""
eval "determinant_$1_formatted=\"d-\""
else
eval "determinant_$1=\"de \""
eval "determinant_$1_formatted=\"de-\""
fi
}
function format () {
eval "$1_formatted=\$(tr --complement --squeeze-repeats 'A-Za-z-_\n' - <<< \"\${$1// /_}\")"
}
function renomme_fichiers () {
for char in $(find characters -type f -regex "characters/[0-9]+-$1.txt"); do
local new_char=$(sed "s/$1/$2/" <<< $char)
echo "character: $char -> $new_char"
mv $char $new_char
break
done
for chapter in $(find outline -type d -regex "outline/.*$1.*"); do
local new_chapter=$(sed "s/$1/$2/g" <<< $chapter)
echo "chapter: $chapter -> $new_chapter"
mv $chapter $new_chapter
done
for part in $(find outline -type f -regex "outline/[^/]*$1[^/]*.md"); do
local new_part=$(sed "s/$1/$2/g" <<< $part)
echo "part: $part -> $new_part"
mv $part $new_part
done
}
trap trap_exit EXIT
temp=$(mktemp --directory /dev/shm/XXXXXXXXX)
cd $temp
if [[ $nouveau = @(check|prox|leven) ]]; then
unzip -qq "$manuscrit"
for param in $(seq 4 $#); do
for dst in ins del rep max; do
eval "if test -n '\$$param' && [[ \"\$$param\" =~ $dst=[0-9]+ ]]; then cost_$dst=\${$param#*=}; fi"
done
done
for dst in ins del rep; do
eval "cost_$dst=\${cost_$dst:-1}"
done
cost_max=${cost_max:-$(($cost_ins + $cost_del + $cost_rep))}
if test $cost_max -ge ${#ancien}; then
cost_max=$(( ${#ancien} - 1 ))
fi
echo paramètres d\'approximation
echo "caractère manquant (del=): $cost_del"
echo "caractère inséré (ins=): $cost_ins"
echo "caractère remplacé (rep=): $cost_rep"
echo "distance maximale (max=): $cost_max"
for f in $(find . -type f); do
let wc+=$(wc -w < $f)
done
awk -v ancien=$ancien -v wc=$wc -v cost_ins=$cost_ins -v cost_del=$cost_del -v cost_rep=$cost_rep -v cost_max=$cost_max '
BEGIN {
RS="[[:punct:]]"
progress_mod = 10
actual_progress = 0
pct_progress = 0
progress = 0
found_words = 0
cost_tot = cost_ins + cost_del + cost_rep
cost_tot = cost_tot > cost_max ? cost_max : cost_tot
str1_len = length(ancien)
for (i=1; i<=str1_len; i++)
str1_substr[i]=substr(ancien, i, 1)
}
function levenshtein(str2) {
str2_len = length(str2)
if(str2_len == 0) return str1_len * cost_del
for(j = 1; j <= str2_len; j++)
str2_substr[j]=substr(str2, j, 1)
matrix[0, 0] = 0
for(i = 1; i <= str1_len; i++) {
matrix[i, 0] = i * cost_del
for(j = 1; j <= str2_len; j++) {
matrix[0, j] = j * cost_ins
x = matrix[i - 1, j] + cost_del
y = matrix[i, j - 1] + cost_ins
z = matrix[i - 1, j - 1] + (str1_substr[i] == str2_substr[j] ? 0 : cost_rep)
x = x < y ? x : y
matrix[i, j] = x < z ? x : z
}
}
return matrix[str1_len, str2_len]
}
{
for (word=1; word<=NF; word++) {
progress++
lvstn = levenshtein(gensub("[[:punct:]]","","g",$word))
if (lvstn <= cost_tot && lvstn > 0) {
key = sprintf("%s (%d)", $word, lvstn)
approx_possibles[key]++
found_words++
}
pct_progress=int(progress / wc * 100)
if (actual_progress < pct_progress && pct_progress % progress_mod == 0) {
actual_progress = pct_progress
printf("%i%\n", actual_progress)
}
}
}
END {
if (found_words > 0) {
pluriel = found_words > 1 ? "s" : ""
printf("mot%s proche%s de «%s» (distance) [occurences]\n", pluriel, pluriel, ancien)
for (i in approx_possibles)
printf("- %s [%i]\n", i, approx_possibles[i])
}
else {
print "aucun mot proche et différent de «" ancien "» trouvé"
}
}
' $(find . -type f)
exit
fi
mv --backup=numbered "$manuscrit" "$backup"
unzip -qq "$backup"
for version in ancien nouveau; do
format $version
determinant $version
done
declare -A remplacement
remplacement=(
[$ancien]="$nouveau"
[$ancien_formatted]="$nouveau_formatted"
[$determinant_ancien$ancien]="$determinant_nouveau$nouveau"
[$determinant_ancien_formatted$ancien_formatted]="$determinant_nouveau_formatted$nouveau_formatted"
)
renomme_fichiers "$ancien_formatted" "$nouveau_formatted"
renomme_fichiers "$determinant_ancien_formatted$ancien_formatted" "$determinant_nouveau_formatted$nouveau_formatted"
egrep --word-regexp --only-matching --recursive --regexp="($determinant_ancien|$determinant_ancien_formatted)\?($ancien|$ancien_formatted)" . \
| awk -v name="$1" -F ':' '
{
if ($NF > 0) {
file[$1]++
nb++
}
}
END {
printf("remplacement de %i occurences pour %s dans %i fichiers\n", nb, name, asort(file))
}'
for regexp in "${!remplacement[@]}"; do
egrep --word-regexp --files-with-matches --recursive --regexp="$regexp" . \
| xargs --no-run-if-empty sed --regexp-extended --in-place "s/(\W|^)$regexp(\W|$)/\1${remplacement[$regexp]}\2/g"
done
zip --recurse-paths --no-dir-entries -qq "${manuscrit}" *