Reorganize file

This commit is contained in:
Michael RICOIS 2014-08-13 13:55:26 +00:00
parent 8c95b1d569
commit e86f75adec
20 changed files with 1731 additions and 0 deletions

90
scripts/build/README Normal file
View File

@ -0,0 +1,90 @@
La base de données doit communiqué en UTF-8
sql_query_pre = SET NAMES utf8
charset_type = utf-8
# 'utf-8' defaults for English and Russian
charset_table = 0..9, A..Z->a..z, _, a..z, \
U+410..U+42F->U+430..U+44F, U+430..U+44F
# For french
charset_table = 0..9, A..Z->a..z, a..z, \
U+00C0..U+00D6->U+00E0..U+00F6, U+00E0..U+00F6, \
U+00D8..U+00DE->U+00F8..U+00FE, U+00F8..U+00FE, \
U+0152->U+0153, U+0153 \
Morphology
==========
# builtin preprocessors are 'none', 'stem_en', 'stem_ru', 'stem_enru',
# 'soundex', and 'metaphone'; additional preprocessors available from
# libstemmer are 'libstemmer_XXX', where XXX is algorithm code
# (see libstemmer_c/libstemmer/modules.txt)
morphology = libstemmer_french
Taille minimum des mots ou l'on applique la morphology
min_stemming_len = 4
According to libstimmer.c/libstimmer/modules.txt, the french module can be refered to by either
french, fr, fre, fra... french UTF_8,ISO_8859_1 french,fr,fre,fra
Utilisation des stopwords
=========================
un, une, le, la, les, de, des, du, dans, l', d', @
Utilisation des wordforms
=========================
Appliquer après les règles de charset_table
Stemming n'est pas appliquer à ces mots , ils deviennent des exceptions
Les mots définis sont utilisé pour normalisé les mots durant l'indexation et la recherche,
il est donc nécessaire de réindexé et redémarrer la recherche pour appliquer les changements
Divers
------
& > ET
Les chiffres
------------
un > 1
deux > 2
trois > 3
quatre > 4
cinq > 5
six > 6
sept > 7
huit > 8
neuf > 9
dix > 10
onze > 11
douze > 12
treize > 13
quatorze > 14
quinze > 15
seize > 16
vingt > 20
vingts > 20
trente > 30
quarante > 40
cinquante > 50
soixante > 60
quatrevingt > 80
cent > 100
cents > 100
mille > 1000
Les chiffres romain
-------------------
Les éléments de voies et abbréviations
--------------------------------------
Voir le wordforms

View File

@ -0,0 +1,42 @@
source act
{
type = mysql
sql_host = 192.168.78.230
sql_user = sphinx
sql_pass = indexer
sql_db = jo
sql_query_pre = SET NAMES utf8
sql_query = \
SELECT l.id, l.idPar, l.idAct, r.siren, r.actif, r.PpPm, r.RS, r.adresse_cp, r.adresse_ville, p.libPays AS pays, l.PDetention \
FROM liens2 l, liensRef r, tabPays p \
WHERE l.dateSuppr = '0000-00-00 00:00:00' \
AND r.id = l.idAct \
AND ( r.siren>1000 OR (r.siren=0 AND r.adresse_pays!='FRA') OR (r.siren=0 AND r.PpPm='PP') ) \
AND p.codPays3 = r.adresse_pays;
sql_attr_uint = idAct
sql_attr_uint = actif
sql_attr_float= PDetention
sql_attr_string = pays
}
index act
{
source = act
path = /dbs/sphinx/act
docinfo = extern
morphology = none
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
}

View File

@ -0,0 +1,157 @@
source ciblage
{
type = mysql
sql_host = 192.168.78.230
sql_user = sphinx
sql_pass = indexer
sql_db = jo
sql_query_pre = SET NAMES utf8
sql_query = \
SELECT id, LPAD(siren, 9, '000000000') AS siren, LPAD(nic, 5, '00000') AS nic, \
siege, \
adr_cp, \
REPLACE(REPLACE(adr_dep, '2B', '202'), '2A', '201') AS adr_dep, \
IF(tel>0,1,0) AS tel, \
IF(fax>0,1,0) AS fax, \
cj, \
capital, \
CONCAT('EX ', ape_etab) AS ape_etab, \
CONCAT('EX ', ape_entrep) AS ape_entrep, \
age_entrep, \
age_etab, \
tca, \
tcaexp, \
IF(teff_entrep IS NULL,99,teff_entrep) AS teff_entrep, \
IF(teff_etab IS NULL,99,teff_etab) AS teff_etab, \
IF(web='',0,1) AS web, \
IF(mail='',0,1) AS mail, \
IF(adrDom>0,1,0) AS adrDom, \
actifEco, \
presentRcs, \
procolHisto, \
tvaIntraValide, \
dateCrea_etab,\
dateCrea_ent, \
dateImmat, \
eff_entrep, \
eff_etab, \
IF(dirNom='',0,1) AS dirNom, \
nbEtab, \
IF(nbMPubli>0,1,0) AS nbMPubli, \
IF(CAST(sirenGrp AS UNSIGNED)>100,1,0) AS sirenGrp, \
nbActio, \
IF(nbActio>0,1,0) AS actio, \
nbPart, \
IF(nbPart>0,1,0) AS part, \
CASE bilType WHEN 'I' THEN 1 WHEN 'R' THEN 2 WHEN 'E' THEN 3 ELSE 0 END as bilType, \
bilAnnee, \
bilCloture, \
bilDuree, \
bilTca, \
bilEE, \
bilFL, \
bilFK, \
bilFR, \
bilGF, \
bilGP, \
bilGU, \
bilGW, \
bilHD, \
bilHH, \
bilHL, \
bilHM, \
bilHN, \
bilYP, \
CAST(codeCommune AS UNSIGNED) AS codeCommune, \
CASE zus WHEN '' THEN 0 WHEN 'HORSZONE' THEN 0 WHEN 'NSP' THEN 0 WHEN 'X' THEN 0 ELSE 1 END as zus, \
CASE zfu WHEN '' THEN 0 WHEN 'HORSZONE' THEN 0 WHEN 'NSP' THEN 0 WHEN 'X' THEN 0 ELSE 1 END as zfu, \
CASE zru WHEN '' THEN 0 WHEN 'NSP' THEN 0 ELSE 1 END as zru, \
CASE cucs WHEN '' THEN 0 WHEN 'NSP' THEN 0 ELSE 1 END as cucs, \
IF(zrr=1,1,0) as zrr, \
IF(zafr=1,1,0) as zafr, \
CASE avisCs WHEN 0 THEN 1 WHEN 10 THEN 2 WHEN 15 THEN 3 WHEN 23 THEN 4 WHEN 29 THEN 4 WHEN 39 THEN 4 WHEN 43 THEN 4 WHEN 21 THEN 5 WHEN 26 THEN 5 WHEN 28 THEN 5 WHEN 31 THEN 6 WHEN 50 THEN 6 WHEN 24 THEN 7 ELSE 0 END as avisCs \
FROM etablissements_act WHERE siren>100;
sql_field_string = siren
sql_field_string = nic
sql_attr_uint = siege
sql_attr_uint = adr_cp
sql_attr_uint = adr_dep
sql_attr_uint = tel
sql_attr_uint = fax
sql_attr_uint = cj
sql_attr_uint = capital
sql_attr_uint = age_entrep
sql_attr_uint = age_etab
sql_attr_uint = tca
sql_attr_uint = tcaexp
sql_attr_uint = teff_entrep
sql_attr_uint = teff_etab
sql_attr_uint = web
sql_attr_uint = mail
sql_attr_uint = adrDom
sql_attr_uint = actifEco
sql_attr_uint = presentRcs
sql_attr_uint = procolHisto
sql_attr_uint = tvaIntraValide
sql_attr_uint = dateCrea_etab
sql_attr_uint = dateCrea_ent
sql_attr_uint = dateImmat
sql_attr_uint = eff_entrep
sql_attr_uint = eff_etab
sql_attr_uint = dirNom
sql_attr_uint = nbEtab
sql_attr_uint = nbMPubli
sql_attr_uint = sirenGrp
sql_attr_uint = nbActio
sql_attr_uint = actio
sql_attr_uint = nbPart
sql_attr_uint = part
sql_attr_uint = bilType
sql_attr_uint = bilAnnee
sql_attr_uint = bilCloture
sql_attr_uint = bilDuree
sql_attr_uint = bilTca
sql_attr_uint = bilEE
sql_attr_uint = bilFL
sql_attr_uint = bilFK
sql_attr_uint = bilFR
sql_attr_uint = bilGF
sql_attr_uint = bilGP
sql_attr_uint = bilGU
sql_attr_uint = bilGW
sql_attr_uint = bilHD
sql_attr_uint = bilHH
sql_attr_uint = bilHL
sql_attr_uint = bilHM
sql_attr_uint = bilHN
sql_attr_uint = bilYP
sql_attr_uint = codeCommune
sql_attr_uint = zus
sql_attr_uint = zru
sql_attr_uint = zfu
sql_attr_uint = cucs
sql_attr_uint = zrr
sql_attr_uint = zafr
sql_attr_uint = avisCs
}
index ciblage
{
source = ciblage
path = /dbs/sphinx/ciblage
docinfo = extern
morphology = none
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
}

View File

@ -0,0 +1,110 @@
#############################################################################
# Documentation
# =============
#
# Présence d'éléments
# IF(element=='',0,1) AS pElement
#
# Liste des champs dans la BDD
# -----------------------------
# id
# source
# source_id
# triCode
# autre_id
# siren => LPAD(siren, 9, '000000000')
# nic => LPAD(nic, 5, '00000')
# actif => 0 ou 1
# siege => 0 ou 1
# raisonSociale
# enseigne
# sigle
# identite_pre
# marques => IF(marques=='',0,1) AS pMarques
# adr_num
# adr_btq
# adr_typeVoie
# adr_libVoie
# adr_comp
# adr_cp => Code postal
# adr_ville => Ville
# adr_dep => Departement (Corse 2A,2B => 201, 202)
# adr_com
# tel => IF(tel>0,1,0) AS pTel
# fax => IF(fax>0,1,0) AS pFax
# cj => Texte
# capital => Float
# capitalDev
# capitalSrc
# ape_etab => Texte => Code APE de l'etablissement
# ape_entrep => Texte => Code APE de l'entreprise
# age_entrep => Entier => Age de l'entreprise
# age_etab => Entier => Age de l'etablissement
# tca => Entier => Tranche de chiffre d'affaire
# tcaexp => Entier => Tranche de chiffre d'affaire à l'export
# teff_entrep => Tranche Effectif de l'entreprise
# teff_etab => Tranche Effectif de l'etablissement
# rang
# web => IF(web=='',0,1) AS pWeb
# mail => IF(mail=='',0,1) AS pMail
# adrDom => 0,1,2
# lieuAct
# actifEco => 0,1
# presentRcs => 0,1
# procolHisto =>
# tvaIntraCle
# tvaIntraValide => 0,1
# ape4_etab
# ape4_entrep
# NaceEtab
# NaceEntrep
# dateCrea_etab => Date
# dateCrea_ent => Date
# dateImmat => Date
# eff_entrep => Entier => Effectif de l'entreprise
# eff_etab => Entier => Effectir de l'etablissement
# distSP
# achPost
# rivoli
# dirCiv
# dirNom => IF(pDirNom=='',0,1) AS pDirNom
# dirPrenom
# dirDateNaiss
# dirFct
# nbEtab => Nombre d'établissement
# nbMPubli
# sirenGrp => IF(pSirenGrp>0,1,0) AS pSirenGrp
# nbActio => Entier => Nombre d'actionnaires (actio,bool)
# nbPart => Entier => Nombre de participations (part, bool)
# bilType =>
# bilAnnee =>
# bilCloture =>
# bilDuree =>
# bilTca =>
# bilEE =>
# bilFL =>
# bilFK =>
# bilFR =>
# bilGF =>
# bilGP =>
# bilGU =>
# bilGW =>
# bilHD =>
# bilHH =>
# bilHL =>
# bilHM =>
# bilHN =>
# bilYP =>
# avisCs
# codeCommune =>
# l93_x
# l93_y
# alt
# precis
# zus =>
# zru =>
# zfu =>
# cucs =>
# zrr =>
# zafr =>
#############################################################################

View File

@ -0,0 +1,64 @@
source dir
{
type = mysql
sql_host = 192.168.78.230
sql_user = sphinx
sql_pass = indexer
sql_db = jo
sql_query_pre = SET NAMES utf8
sql_query = \
SELECT id, siren, adr_dep, typeDir, dirSiren, civilite, CONCAT(nom,' ',naissance_nom, ' ', dirRS) AS nom, prenom, \
YEAR(naissance_date) AS naiss_annee, \
MONTH(naissance_date) AS naiss_mois, \
DAY(naissance_date) AS naiss_jour, \
naissance_lieu, fonction_code, actif \
FROM dirigeants;
sql_attr_string = civilite
sql_attr_string = typeDir
sql_attr_string = fonction_code
sql_attr_uint = dirSiren
sql_attr_uint = naiss_annee
sql_attr_uint = naiss_mois
sql_attr_uint = naiss_jour
sql_attr_uint = actif
sql_attr_uint = adr_dep
}
index dir
{
source = dir
path = /dbs/sphinx/dir
docinfo = extern
morphology = none
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
}
index dir_phx
{
source = dir
path = /dbs/sphinx/dir_phx
docinfo = extern
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
min_stemming_len = 4
}

View File

@ -0,0 +1,63 @@
source dir
{
type = mysql
sql_host = 192.168.78.230
sql_user = sphinx
sql_pass = indexer
sql_db = jo
sql_query_pre =
sql_query = \
SELECT id, siren, adr_dep, typeDir, dirSiren, civilite, CONCAT(nom,' ',naissance_nom, ' ', dirRS) AS nom, prenom, \
YEAR(naissance_date) AS naiss_annee, \
MONTH(naissance_date) AS naiss_mois, \
DAY(naissance_date) AS naiss_jour, \
naissance_lieu, fonction_code, actif \
FROM dirigeants_tmp;
sql_attr_string = civilite
sql_attr_string = typeDir
sql_attr_string = fonction_code
sql_attr_uint = dirSiren
sql_attr_uint = naiss_annee
sql_attr_uint = naiss_mois
sql_attr_uint = naiss_jour
sql_attr_uint = actif
sql_attr_uint = adr_dep
}
index dir
{
source = dir
path = /dbs/sphinx/dir
docinfo = extern
morphology = none
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
}
index dir_phx
{
source = dir
path = /dbs/sphinx/dir_phx
docinfo = extern
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
min_stemming_len = 4
}

View File

@ -0,0 +1,68 @@
source ent
{
type = mysql
sql_host = 192.168.78.230
sql_user = sphinx
sql_pass = indexer
sql_db = jo
sql_query_pre = SET NAMES utf8
sql_query = SELECT id, siren, nic, siege, \
CONCAT_WS(" ", raisonSociale, enseigne, sigle, identite_pre) AS nom, \
REPLACE(REPLACE(adr_dep, '2B', '202'), '2A', '201') AS adr_dep, \
actif, adr_num, CONCAT_WS(" ", adr_typeVoie, adr_libVoie ,adr_comp) AS adresse, adr_cp, \
adr_ville AS ville, cj, ape_etab, IF(siren>200,1,0) AS sirenValide, rang \
FROM etablissements;
sql_attr_uint = siren
sql_attr_uint = nic
sql_attr_uint = siege
sql_attr_uint = actif
sql_attr_uint = adr_num
sql_attr_uint = adr_cp
sql_attr_uint = adr_dep
sql_attr_uint = cj
sql_attr_uint = sirenValide
sql_attr_uint = rang
}
index ent
{
source = ent
path = /dbs/sphinx/ent
docinfo = extern
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
wordforms = /etc/sphinxsearch/wordforms-ent.txt
enable_star = 1
}
index ent_phx
{
source = ent
path = /dbs/sphinx/ent_phx
docinfo = extern
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
wordforms = /usr/local/sphinx/etc/wordforms-ent.txt
enable_star = 1
morphology = libstemmer_fr
min_stemming_len = 4
}

View File

@ -0,0 +1,69 @@
source ent
{
type = mysql
sql_host = 192.168.78.230
sql_user = sphinx
sql_pass = indexer
sql_db = jo
sql_query_pre = SET NAMES utf8
sql_query = SELECT id, siren, nic, siege, \
CONCAT_WS(" ", raisonSociale, enseigne, sigle, identite_pre) AS nom, \
REPLACE(REPLACE(adr_dep, '2B', '202'), '2A', '201') AS adr_dep, \
actif, adr_num, CONCAT_WS(" ", adr_typeVoie, adr_libVoie ,adr_comp) AS adresse, adr_cp, \
adr_ville AS ville, cj, ape_etab, IF(siren>200,1,0) AS sirenValide, rang \
FROM etablissements_tmp;
sql_attr_uint = siren
sql_attr_uint = nic
sql_attr_uint = siege
sql_attr_uint = actif
sql_attr_uint = adr_num
sql_attr_uint = adr_cp
sql_attr_uint = adr_dep
sql_attr_uint = cj
sql_attr_uint = sirenValide
sql_attr_uint = rang
}
index ent
{
source = ent
path = /dbs/sphinx/ent
docinfo = extern
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
wordforms = /etc/sphinxsearch/wordforms-ent.txt
enable_star = 1
}
index ent_phx
{
source = ent
path = /dbs/sphinx/ent_phx
docinfo = extern
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
wordforms = /usr/local/sphinx/etc/wordforms-ent.txt
enable_star = 1
morphology = libstemmer_fr
min_stemming_len = 4
}

View File

@ -0,0 +1,24 @@
source histo
{
type = mysql
sql_host = 192.168.78.230
sql_user = sphinx
sql_pass = indexer
sql_db = histobodacc
sql_port = 3306 # optional, default is 3306
sql_query = SELECT id, nomFichier, annee1, dateBod, texte FROM bodacc_ocr;
sql_attr_uint = annee1
sql_query_info = SELECT * FROM bodacc_ocr WHERE id=$id
}
index histo
{
source = histo
path = /dbs/sphinx/histo
docinfo = extern
morphology = none
min_word_len = 2
charset_type = sbcs
html_strip = 1
}

View File

@ -0,0 +1,79 @@
#############################################################################
## indexer settings
#############################################################################
indexer
{
# memory limit, in bytes, kiloytes (16384K) or megabytes (256M)
# optional, default is 32M, max is 2047M, recommended is 256M to 1024M
mem_limit = 256M
# maximum IO calls per second (for I/O throttling)
# optional, default is 0 (unlimited)
#
# max_iops = 40
# maximum IO call size, bytes (for I/O throttling)
# optional, default is 0 (unlimited)
#
# max_iosize = 1048576
}
#############################################################################
## searchd settings
#############################################################################
searchd
{
# IP address to bind on
# optional, default is 0.0.0.0 (ie. listen on all interfaces)
#
# address = 127.0.0.1
# address = 192.168.0.1
# searchd TCP port number
# mandatory, default is 3312
listen = 3312
# log file, searchd run info is logged here
# optional, default is 'searchd.log'
log = /dbs/sphinxlog/searchd.log
# query log file, all search queries are logged here
# optional, default is empty (do not log queries)
query_log = /dbs/sphinxlog/query.log
# client read timeout, seconds
# optional, default is 5
read_timeout = 5
# maximum amount of children to fork (concurrent searches to run)
# optional, default is 0 (unlimited)
max_children = 30
# PID file, searchd process ID file name
# mandatory
pid_file = /var/log/searchd.pid
# max amount of matches the daemon ever keeps in RAM, per-index
# WARNING, THERE'S ALSO PER-QUERY LIMIT, SEE SetLimits() API CALL
# default is 1000 (just like Google)
max_matches = 1000
# seamless rotate, prevents rotate stalls if precaching huge datasets
# optional, default is 1
seamless_rotate = 1
# whether to forcibly preopen all indexes on startup
# optional, default is 0 (do not preopen)
preopen_indexes = 1
# whether to unlink .old index copies on succesful rotation.
# optional, default is 1 (do unlink)
unlink_old = 1
compat_sphinxql_magics=0
}

View File

@ -0,0 +1,207 @@
& > et
un > 1
deux > 2
trois > 3
quatre > 4
cinq > 5
six > 6
sept > 7
huit > 8
neuf > 9
dix > 10
onze > 11
douze > 12
treize > 13
quatorze > 14
quinze > 15
seize > 16
vingt > 20
vingts > 20
trente > 30
quarante > 40
cinquante > 50
soixante > 60
quatrevingt > 80
cent > 100
cents > 100
mille > 1000
zac > zone
zad > zone
za > zone
zi > zone
zup > zone
general > gal
abbaye > abe
agglomeration > agl
aglo > agl
allee > all
ancien > ach
ancienne > art
anse > anse
arcade > arc
autoroute > aut
avenue > av
barriere > bre
bas > bch
bastide > bstd
baston > bast
beguinage > begi
berge > ber
bois > bois
boite postal > bp
boucle > bcle
boulevard > bd
bourg > brg
butte > but
campagne > cgne
camping > cpg
carre > carr
carreau > cau
carrefour > car
carriere > care
castel > cst
cavee > cav
central > ctre
centre > ctre
chalet > chl
chapelle > chp
charmille > chi
chateau > cht
chaussee > chs
che > chemin
chv > chemin
cheminement > chem
cloitre > cloi
colline > coli
contour > ctr
corniche > cor
cottage > cott
cours > crs
darse > dars
degre > deg
dsg > descente
dsc > descente
digue > dig
domaine > dom
docteur > dr
ecart > eca
ecluse > ecl
eglise > egl
enceinte > en
enclave > env
enclos > enc
escalier > esc
espace > espa
esplanade > esp
etang > eting
faubourg > fg
ferme > frm
fontaine > fon
fort > fort
forum > form
fosse > fos
foyer > foyr
galerie > gal
garenne > garn
grand > gbd
gden > grande
gr > grande
grille > gri
grimpette > grim
groupe > gpe
groupement > gpt
halle > hle
hameau > ham
haut > hch
hippodrome > hip
immeuble > imm
impasse > imp
jardin > jard
jetee > jte
levee > leve
lieu > ld
lieudit > ld
lotissement > lot
maison > mf
manoir > man
marche > mar
metro > met
montee > mte
moulin > mln
musee > mus
nouvelle > nte
palais > pal
parc > parc
parking > pkg
parvis > prv
pas > passage
pn > passage
passe > pass
passerelle > ple
patio > pat
pavillon > pav
peripherique > peri
peristyle > psty
petites > pta
pae > petite
pim > petite
prt > petite
ptr > petite
place > pl
placis > plci
plage > plag
plaine > pln
plan > plan
plateau > plt
pointe > pnt
porche > pch
porte > pte
portique > porq
poterne > pot
pourtour > pour
presqu ile > prq
promenade > prom
quai > qu
quartier > qua
raccourci > rac
raidillon > raid
rampe > rpe
rempart > rem
residence > res
rocade > roc
rond > rpt
roquet > roqt
rotonde > rtd
route > rte
rue > r
ruelle > rle
sente > sen
sentier > sen
square > sq
st > saint
ste > saint
sainte > saint
stade > stde
station > sta
terre > tpl
terrain > trn
terrasse > tsse
tertre > trt
traverse > tra
vallon > val
vallee > val
venelle > ven
vieille > vte
villa > vla
vge > village
vlge > village
voie > voi
centre cial > ccal
centre com > ccal
centre comm > ccal
centre commercial > ccal
ville > mairie
commune > mairie
conseil regional > region
conseil general > departement
companie > cie

View File

@ -0,0 +1,42 @@
source act
{
type = mysql
sql_host = 192.168.78.230
sql_user = sphinx
sql_pass = indexer
sql_db = jo
sql_query_pre = SET NAMES utf8
sql_query = \
SELECT l.id, l.idPar, l.idAct, r.siren, r.actif, r.PpPm, r.RS, r.adresse_cp, r.adresse_ville, p.libPays AS pays, l.PDetention \
FROM liens2 l, liensRef r, tabPays p \
WHERE l.dateSuppr = '0000-00-00 00:00:00' \
AND r.id = l.idAct \
AND ( r.siren>1000 OR (r.siren=0 AND r.adresse_pays!='FRA') OR (r.siren=0 AND r.PpPm='PP') ) \
AND p.codPays3 = r.adresse_pays;
sql_attr_uint = idAct
sql_attr_uint = actif
sql_attr_float= PDetention
sql_attr_string = pays
}
index act
{
source = act
path = /dbs/sphinx/act
docinfo = extern
morphology = none
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
}

View File

@ -0,0 +1,157 @@
source ciblage
{
type = mysql
sql_host = 192.168.3.30
sql_user = sphinx
sql_pass = indexer
sql_db = jo
sql_query_pre = SET NAMES utf8
sql_query = \
SELECT id, LPAD(siren, 9, '000000000') AS siren, LPAD(nic, 5, '00000') AS nic, \
siege, \
adr_cp, \
REPLACE(REPLACE(adr_dep, '2B', '202'), '2A', '201') AS adr_dep, \
IF(tel>0,1,0) AS tel, \
IF(fax>0,1,0) AS fax, \
cj, \
capital, \
CONCAT('EX ', ape_etab) AS ape_etab, \
CONCAT('EX ', ape_entrep) AS ape_entrep, \
age_entrep, \
age_etab, \
tca, \
tcaexp, \
IF(teff_entrep IS NULL,99,teff_entrep) AS teff_entrep, \
IF(teff_etab IS NULL,99,teff_etab) AS teff_etab, \
IF(web='',0,1) AS web, \
IF(mail='',0,1) AS mail, \
IF(adrDom>0,1,0) AS adrDom, \
actifEco, \
presentRcs, \
procolHisto, \
tvaIntraValide, \
dateCrea_etab,\
dateCrea_ent, \
dateImmat, \
eff_entrep, \
eff_etab, \
IF(dirNom='',0,1) AS dirNom, \
nbEtab, \
IF(nbMPubli>0,1,0) AS nbMPubli, \
IF(CAST(sirenGrp AS UNSIGNED)>100,1,0) AS sirenGrp, \
nbActio, \
IF(nbActio>0,1,0) AS actio, \
nbPart, \
IF(nbPart>0,1,0) AS part, \
CASE bilType WHEN 'I' THEN 1 WHEN 'R' THEN 2 WHEN 'E' THEN 3 ELSE 0 END as bilType, \
bilAnnee, \
bilCloture, \
bilDuree, \
bilTca, \
bilEE, \
bilFL, \
bilFK, \
bilFR, \
bilGF, \
bilGP, \
bilGU, \
bilGW, \
bilHD, \
bilHH, \
bilHL, \
bilHM, \
bilHN, \
bilYP, \
CAST(codeCommune AS UNSIGNED) AS codeCommune, \
CASE zus WHEN '' THEN 0 WHEN 'HORSZONE' THEN 0 WHEN 'NSP' THEN 0 WHEN 'X' THEN 0 ELSE 1 END as zus, \
CASE zfu WHEN '' THEN 0 WHEN 'HORSZONE' THEN 0 WHEN 'NSP' THEN 0 WHEN 'X' THEN 0 ELSE 1 END as zfu, \
CASE zru WHEN '' THEN 0 WHEN 'NSP' THEN 0 ELSE 1 END as zru, \
CASE cucs WHEN '' THEN 0 WHEN 'NSP' THEN 0 ELSE 1 END as cucs, \
IF(zrr=1,1,0) as zrr, \
IF(zafr=1,1,0) as zafr, \
CASE avisCs WHEN 0 THEN 1 WHEN 10 THEN 2 WHEN 15 THEN 3 WHEN 23 THEN 4 WHEN 29 THEN 4 WHEN 39 THEN 4 WHEN 43 THEN 4 WHEN 21 THEN 5 WHEN 26 THEN 5 WHEN 28 THEN 5 WHEN 31 THEN 6 WHEN 50 THEN 6 WHEN 24 THEN 7 ELSE 0 END as avisCs \
FROM etablissements_act WHERE siren>100;
sql_field_string = siren
sql_field_string = nic
sql_attr_uint = siege
sql_attr_uint = adr_cp
sql_attr_uint = adr_dep
sql_attr_uint = tel
sql_attr_uint = fax
sql_attr_uint = cj
sql_attr_uint = capital
sql_attr_uint = age_entrep
sql_attr_uint = age_etab
sql_attr_uint = tca
sql_attr_uint = tcaexp
sql_attr_uint = teff_entrep
sql_attr_uint = teff_etab
sql_attr_uint = web
sql_attr_uint = mail
sql_attr_uint = adrDom
sql_attr_uint = actifEco
sql_attr_uint = presentRcs
sql_attr_uint = procolHisto
sql_attr_uint = tvaIntraValide
sql_attr_uint = dateCrea_etab
sql_attr_uint = dateCrea_ent
sql_attr_uint = dateImmat
sql_attr_uint = eff_entrep
sql_attr_uint = eff_etab
sql_attr_uint = dirNom
sql_attr_uint = nbEtab
sql_attr_uint = nbMPubli
sql_attr_uint = sirenGrp
sql_attr_uint = nbActio
sql_attr_uint = actio
sql_attr_uint = nbPart
sql_attr_uint = part
sql_attr_uint = bilType
sql_attr_uint = bilAnnee
sql_attr_uint = bilCloture
sql_attr_uint = bilDuree
sql_attr_uint = bilTca
sql_attr_uint = bilEE
sql_attr_uint = bilFL
sql_attr_uint = bilFK
sql_attr_uint = bilFR
sql_attr_uint = bilGF
sql_attr_uint = bilGP
sql_attr_uint = bilGU
sql_attr_uint = bilGW
sql_attr_uint = bilHD
sql_attr_uint = bilHH
sql_attr_uint = bilHL
sql_attr_uint = bilHM
sql_attr_uint = bilHN
sql_attr_uint = bilYP
sql_attr_uint = codeCommune
sql_attr_uint = zus
sql_attr_uint = zru
sql_attr_uint = zfu
sql_attr_uint = cucs
sql_attr_uint = zrr
sql_attr_uint = zafr
sql_attr_uint = avisCs
}
index ciblage
{
source = ciblage
path = /dbs/sphinx/ciblage
docinfo = extern
morphology = none
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
}

View File

@ -0,0 +1,64 @@
source dir
{
type = mysql
sql_host = 192.168.78.230
sql_user = sphinx
sql_pass = indexer
sql_db = jo
sql_query_pre = SET NAMES utf8
sql_query = \
SELECT id, siren, adr_dep, typeDir, dirSiren, civilite, CONCAT(nom,' ',naissance_nom, ' ', dirRS) AS nom, prenom, \
YEAR(naissance_date) AS naiss_annee, \
MONTH(naissance_date) AS naiss_mois, \
DAY(naissance_date) AS naiss_jour, \
naissance_lieu, fonction_code, actif \
FROM dirigeants;
sql_attr_string = civilite
sql_attr_string = typeDir
sql_attr_string = fonction_code
sql_attr_uint = dirSiren
sql_attr_uint = naiss_annee
sql_attr_uint = naiss_mois
sql_attr_uint = naiss_jour
sql_attr_uint = actif
sql_attr_uint = adr_dep
}
index dir
{
source = dir
path = /dbs/sphinx/dir
docinfo = extern
morphology = none
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
}
index dir_phx
{
source = dir
path = /dbs/sphinx/dir_phx
docinfo = extern
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
min_stemming_len = 4
}

View File

@ -0,0 +1,63 @@
source dir
{
type = mysql
sql_host = 192.168.78.230
sql_user = sphinx
sql_pass = indexer
sql_db = jo
sql_query_pre =
sql_query = \
SELECT id, siren, adr_dep, typeDir, dirSiren, civilite, CONCAT(nom,' ',naissance_nom, ' ', dirRS) AS nom, prenom, \
YEAR(naissance_date) AS naiss_annee, \
MONTH(naissance_date) AS naiss_mois, \
DAY(naissance_date) AS naiss_jour, \
naissance_lieu, fonction_code, actif \
FROM dirigeants_tmp;
sql_attr_string = civilite
sql_attr_string = typeDir
sql_attr_string = fonction_code
sql_attr_uint = dirSiren
sql_attr_uint = naiss_annee
sql_attr_uint = naiss_mois
sql_attr_uint = naiss_jour
sql_attr_uint = actif
sql_attr_uint = adr_dep
}
index dir
{
source = dir
path = /dbs/sphinx/dir
docinfo = extern
morphology = none
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
}
index dir_phx
{
source = dir
path = /dbs/sphinx/dir_phx
docinfo = extern
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
min_stemming_len = 4
}

View File

@ -0,0 +1,73 @@
source ent
{
type = mysql
sql_host = 192.168.3.30
sql_user = sphinx
sql_pass = indexer
sql_db = jo
sql_query_pre = SET NAMES utf8
sql_query = SELECT id, siren, nic, siege, \
CONCAT_WS(" ", \
raisonSociale, \
enseigne, \
sigle, \
identite_pre\
) AS nom, REPLACE(REPLACE(adr_dep, '2B', '202'), '2A', '201') AS adr_dep, \
actif, adr_num, CONCAT_WS(" ", adr_typeVoie, adr_libVoie ,adr_comp) AS adresse, adr_cp, \
adr_ville AS ville, cj, ape_etab, (siren>200) AS sirenValide, rang \
FROM etablissements;
sql_attr_uint = siren
sql_attr_uint = nic
sql_attr_uint = siege
sql_attr_uint = actif
sql_attr_uint = adr_num
sql_attr_uint = adr_cp
sql_attr_uint = adr_dep
sql_attr_uint = cj
sql_attr_uint = sirenValide
sql_attr_uint = rang
}
index ent
{
source = ent
path = /dbs/sphinx/ent
docinfo = extern
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
wordforms = /etc/sphinxsearch/wordforms-ent.txt
enable_star = 1
}
index ent_phx
{
source = ent
path = /dbs/sphinx/ent_phx
docinfo = extern
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
wordforms = /usr/local/sphinx/etc/wordforms-ent.txt
enable_star = 1
morphology = libstemmer_fr
min_stemming_len = 4
}

View File

@ -0,0 +1,73 @@
source ent
{
type = mysql
sql_host = 192.168.3.30
sql_user = sphinx
sql_pass = indexer
sql_db = jo
sql_query_pre = SET NAMES utf8
sql_query = SELECT id, siren, nic, siege, \
CONCAT_WS(" ", \
raisonSociale, \
enseigne, \
sigle, \
identite_pre\
) AS nom, REPLACE(REPLACE(adr_dep, '2B', '202'), '2A', '201') AS adr_dep, \
actif, adr_num, CONCAT_WS(" ", adr_typeVoie, adr_libVoie ,adr_comp) AS adresse, adr_cp, \
adr_ville AS ville, cj, ape_etab, (siren>200) AS sirenValide, rang \
FROM etablissements_tmp;
sql_attr_uint = siren
sql_attr_uint = nic
sql_attr_uint = siege
sql_attr_uint = actif
sql_attr_uint = adr_num
sql_attr_uint = adr_cp
sql_attr_uint = adr_dep
sql_attr_uint = cj
sql_attr_uint = sirenValide
sql_attr_uint = rang
}
index ent
{
source = ent
path = /dbs/sphinx/ent
docinfo = extern
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
wordforms = /etc/sphinxsearch/wordforms-ent.txt
enable_star = 1
}
index ent_phx
{
source = ent
path = /dbs/sphinx/ent_phx
docinfo = extern
charset_type = utf-8
charset_table = 0..9, A..Z->a..z, a..z, \
U+23, U+25, U+26, U+2B, U+3D, U+40, \
U+C0->a, U+C1->a, U+C2->a, U+C3->a, U+C4->a, U+C5->a, U+C6->a, U+C7->c, \
U+C8->e, U+C8->e, U+C9->e, U+CA->e, U+CB->e, U+CC->i, U+CD->i, U+CE->i, \
U+CF->i, U+D0->d, U+D1->n, U+D2->o, U+D3->o, U+D4->o, U+D5->o, U+D6->o, \
U+D8->o, U+D9->u, U+DA->u, U+DB->u, U+DC->u, U+DD->y, U+E0->a, U+E1->a, \
U+E2->a, U+E3->a, U+E4->a, U+E5->a, U+E7->c, U+E8->e, U+E9->e, U+EA->e, \
U+EB->e, U+EC->i, U+ED->i, U+EE->i, U+EF->i, U+F1->n, U+F2->o, U+F3->o, \
U+F4->o, U+F5->o, U+F6->o, U+F8->o, U+F9->u, U+FA->u,U+FB->u, U+FC->u, \
U+FD->y, U+FF->y, U+0152->U+0153, U+0153
wordforms = /usr/local/sphinx/etc/wordforms-ent.txt
enable_star = 1
morphology = libstemmer_fr
min_stemming_len = 4
}

View File

@ -0,0 +1,79 @@
#############################################################################
## indexer settings
#############################################################################
indexer
{
# memory limit, in bytes, kiloytes (16384K) or megabytes (256M)
# optional, default is 32M, max is 2047M, recommended is 256M to 1024M
mem_limit = 256M
# maximum IO calls per second (for I/O throttling)
# optional, default is 0 (unlimited)
#
# max_iops = 40
# maximum IO call size, bytes (for I/O throttling)
# optional, default is 0 (unlimited)
#
# max_iosize = 1048576
}
#############################################################################
## searchd settings
#############################################################################
searchd
{
# IP address to bind on
# optional, default is 0.0.0.0 (ie. listen on all interfaces)
#
# address = 127.0.0.1
# address = 192.168.0.1
# searchd TCP port number
# mandatory, default is 3312
listen = 3312
# log file, searchd run info is logged here
# optional, default is 'searchd.log'
log = /dbs/sphinxlog/searchd.log
# query log file, all search queries are logged here
# optional, default is empty (do not log queries)
query_log = /dbs/sphinxlog/query.log
# client read timeout, seconds
# optional, default is 5
read_timeout = 5
# maximum amount of children to fork (concurrent searches to run)
# optional, default is 0 (unlimited)
max_children = 30
# PID file, searchd process ID file name
# mandatory
pid_file = /var/log/searchd.pid
# max amount of matches the daemon ever keeps in RAM, per-index
# WARNING, THERE'S ALSO PER-QUERY LIMIT, SEE SetLimits() API CALL
# default is 1000 (just like Google)
max_matches = 1000
# seamless rotate, prevents rotate stalls if precaching huge datasets
# optional, default is 1
seamless_rotate = 1
# whether to forcibly preopen all indexes on startup
# optional, default is 0 (do not preopen)
preopen_indexes = 1
# whether to unlink .old index copies on succesful rotation.
# optional, default is 1 (do unlink)
unlink_old = 1
compat_sphinxql_magics=0
}

View File

@ -0,0 +1,207 @@
& > et
un > 1
deux > 2
trois > 3
quatre > 4
cinq > 5
six > 6
sept > 7
huit > 8
neuf > 9
dix > 10
onze > 11
douze > 12
treize > 13
quatorze > 14
quinze > 15
seize > 16
vingt > 20
vingts > 20
trente > 30
quarante > 40
cinquante > 50
soixante > 60
quatrevingt > 80
cent > 100
cents > 100
mille > 1000
zac > zone
zad > zone
za > zone
zi > zone
zup > zone
general > gal
abbaye > abe
agglomeration > agl
aglo > agl
allee > all
ancien > ach
ancienne > art
anse > anse
arcade > arc
autoroute > aut
avenue > av
barriere > bre
bas > bch
bastide > bstd
baston > bast
beguinage > begi
berge > ber
bois > bois
boite postal > bp
boucle > bcle
boulevard > bd
bourg > brg
butte > but
campagne > cgne
camping > cpg
carre > carr
carreau > cau
carrefour > car
carriere > care
castel > cst
cavee > cav
central > ctre
centre > ctre
chalet > chl
chapelle > chp
charmille > chi
chateau > cht
chaussee > chs
che > chemin
chv > chemin
cheminement > chem
cloitre > cloi
colline > coli
contour > ctr
corniche > cor
cottage > cott
cours > crs
darse > dars
degre > deg
dsg > descente
dsc > descente
digue > dig
domaine > dom
docteur > dr
ecart > eca
ecluse > ecl
eglise > egl
enceinte > en
enclave > env
enclos > enc
escalier > esc
espace > espa
esplanade > esp
etang > eting
faubourg > fg
ferme > frm
fontaine > fon
fort > fort
forum > form
fosse > fos
foyer > foyr
galerie > gal
garenne > garn
grand > gbd
gden > grande
gr > grande
grille > gri
grimpette > grim
groupe > gpe
groupement > gpt
halle > hle
hameau > ham
haut > hch
hippodrome > hip
immeuble > imm
impasse > imp
jardin > jard
jetee > jte
levee > leve
lieu > ld
lieudit > ld
lotissement > lot
maison > mf
manoir > man
marche > mar
metro > met
montee > mte
moulin > mln
musee > mus
nouvelle > nte
palais > pal
parc > parc
parking > pkg
parvis > prv
pas > passage
pn > passage
passe > pass
passerelle > ple
patio > pat
pavillon > pav
peripherique > peri
peristyle > psty
petites > pta
pae > petite
pim > petite
prt > petite
ptr > petite
place > pl
placis > plci
plage > plag
plaine > pln
plan > plan
plateau > plt
pointe > pnt
porche > pch
porte > pte
portique > porq
poterne > pot
pourtour > pour
presqu ile > prq
promenade > prom
quai > qu
quartier > qua
raccourci > rac
raidillon > raid
rampe > rpe
rempart > rem
residence > res
rocade > roc
rond > rpt
roquet > roqt
rotonde > rtd
route > rte
rue > r
ruelle > rle
sente > sen
sentier > sen
square > sq
st > saint
ste > saint
sainte > saint
stade > stde
station > sta
terre > tpl
terrain > trn
terrasse > tsse
tertre > trt
traverse > tra
vallon > val
vallee > val
venelle > ven
vieille > vte
villa > vla
vge > village
vlge > village
voie > voi
centre cial > ccal
centre com > ccal
centre comm > ccal
centre commercial > ccal
ville > mairie
commune > mairie
conseil regional > region
conseil general > departement
companie > cie