Ajout config et explication ciblage

This commit is contained in:
Michael RICOIS 2012-04-26 14:58:12 +00:00
parent 70c67fa13f
commit 3044222751

View File

@ -0,0 +1,207 @@
#############################################################################
# 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=='0000000000',0,1) AS pTel
# fax => IF(fax=='0000000000',0,1) 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 => nbEtab
# nbMPubli
# sirenGrp => IF(pSirenGrp=='0',0,1) AS pSirenGrp
# nbActio =>
# nbPart =>
# 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 =>
#############################################################################
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 LPAD(siren, 9, '000000000') AS siren, LPAD(nic, 5, '00000') as nic \
actif, siege, \
adr_dep, \
IF(tel=='0000000000',0,1) AS tel, \
IF(fax=='0000000000',0,1) AS fax, \
cj, \
capital, \
ape_etab, \
ape_entrep, \
age_entrep, \
age_etab, \
tca, \
tcaexp, \
teff_entrep, \
teff_etab, \
IF(web=='',0,1) AS pWeb, \
IF(mail=='',0,1) AS pMail, \
actifEco, \
presentRcs, \
tvaIntraValide, \
dateCrea_etab,\
dateCrea_ent, \
dateImmat, \
eff_entrep, \
eff_etab, \
FROM etablissements_act WHERE siren>100;
sql_attr_string = siren
sql_attr_string = nic
sql_attr_uint = actif
sql_attr_uint = siege
sql_attr_uint = adr_dep
sql_attr_uint = tel
sql_attr_uint = fax
sql_attr_uint = cj
sql_attr_uint = capital
sql_attr_str2ordinal = ape_etab
sql_attr_str2ordinal = ape_entrep
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 = actifEco
sql_attr_uint = presentRcs
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
}
SELECT id, actif, siege, siren , nic, adr_cp, REPLACE(REPLACE(adr_dep, '2B', 200), '2A', 200) AS adr_dep,\
adr_com, (tel > 0) AS tel, (fax > 0) AS fax, cj, capital, ape_etab,\
ape_entrep, age_entrep, age_etab, tcaexp, teff_entrep, teff_etab, rang, actifEco,\
presentRcs, procolHisto, tvaIntraValide, NaceEtab, NaceEntrep, dateCrea_etab, dateCrea_ent,\
dateImmat, eff_entrep, eff_etab, rivoli, nbEtab, if(sirenGrp>0, 1, 0) as sirenGrp, nbActio,\
nbPart, bilType, bilAnnee, bilTca, bilEE, bilFL, bilFK, bilFR, bilGF, bilGP, bilGU,\
bilGW, bilHD, bilHH, bilHL, bilHM, bilHN, bilYP, avisCs,\
bilFL - bilFK AS bilFS, bilFR - bilGF AS bilGG, bilGP - bilGU AS bilGV, bilHD - bilHH AS bilHI FROM etablissements_act WHERE siren > 100
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
}