Utilisation du PIVOT avec paramètres inconnus

[justinedr71]

Bonjour,

Je souhaite réaliser la requête suivante :

Code sql :

1
2
3
4
5
6
7
8
9
10
SELECT *
FROM   (SELECT pol_numpol, sor_ident, ccp_libelle, gad_prime_nette
FROM f_polices, f_sit_objet_risque, f_garantie_dyn, t_cdecpt, f_tarif_ass
WHERE gad_ptrsorid = sor_ident
AND sor_ptrpolid = pol_ident
AND sor_datetarif = tas_dateffet
AND tas_codeintercalaire = ccp_code
AND gad_prime_nette IS NOT NULL AND gad_prime_nette != 0
PIVOT XML (SUM(gad_prime_nette) AS prime FOR (ccp_libelle) IN (SELECT DISTINCT ccp_code FROM   t_cdecpt))                                                   
ORDER BY pol_numpol, sor_ident

sans que mon résultat ne soit du XML. Le problème est que je ne peux pas connaitre à l'avance ni le nombre, ni les libellés de mes ccp_code... Comment faire?

[laurentschneider]

dans sql il est impossible d'avoir un nombre dynamique de colonne.

tu peux tenter du plsql

Code :

1
2
3
4
5
6
7
8
9
10
11
12
var r refcursor
 
declare joblist varchar2(4000);
begin
  SELECT listagg( ''''||job||''' as '||job,',') within GROUP (ORDER BY job) 
    INTO joblist FROM (SELECT DISTINCT job FROM emp);
  open :r FOR 'select * from (select deptno,job from emp) pivot  
    (count(*) for (job) in ('||joblist||'))';
end;
/
 
print r

Code :

1
2
3
4
5
    DEPTNO    ANALYST      CLERK    MANAGER  PRESIDENT   SALESMAN
---------- ---------- ---------- ---------- ---------- ----------
        30          0          1          1          0          4
        20          2          2          1          0          0
        10          0          1          1          1          0

[justinedr71]

Merci beaucoup!!!

Donc avec ma requête, si j'ai bien compris ça donne ça :

Code sql :

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
var r refcursor
 
declare joblist varchar2(4000);
 
begin
  SELECT listagg( ''''||ccp_libelle||''' as '||ccp_libelle,',') within GROUP (ORDER BY ccp_libelle)
    INTO joblist FROM (SELECT DISTINCT ccp_libelle FROM t_cdecpt);
  open :r FOR 'select * from (select pol_numpol, sor_ident, ccp_libelle, gad_prime_nette
        from f_polices, f_sit_objet_risque, f_garantie_dyn, t_cdecpt, f_tarif_ass
        where gad_ptrsorid = sor_ident
        and sor_ptrpolid = pol_ident
        and sor_datetarif = tas_dateffet
        and tas_codeintercalaire = ccp_code
        and gad_prime_nette is not null and gad_prime_nette != 0) 
        pivot (count(*) for (ccp_libelle) in ('||joblist||'))';
end;
/
 
print r

il me dit que j'ai un problème de virgule à la ligne 6 lorsque je l'exécute sous TOAD... ???

[justinedr71]

Même en ne mettant que ce code la :

Code :

1
2
3
4
5
6
7
8
9
10
var r refcursor
declare joblist varchar2(4000);
begin
  SELECT listagg( ''''||ccp_libelle||''' as '||ccp_libelle,',') within GROUP (ORDER BY ccp_libelle) 
    INTO joblist FROM (SELECT DISTINCT ccp_libelle FROM t_cdecpt);
    open :r FOR 'select * from (select ccp_code,ccp_libelle from t_cdecpt) pivot  
    (count(*) for (ccp_libelle) in ('||joblist||'))';
end;
/
print r

J'ai toujours un problème de 'virgule' à la ligne 5... J'ai essayé de placer des virgules à plusieurs endroits... Ça ne résout rien

[jkofr]

Aurai-tu une virgule dans la colonne ccp_libelle ?

Jko

[justinedr71]

Non non, juste "ccp_libelle"

[jkofr]

Hello,

Je parlais du contenu de ta colonne ccp_libelle.

Démonstration:

Code :

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
 
var r refcursor
declare joblist varchar2(4000);
begin
  SELECT listagg( ''''||job||''' as '||job,',') within GROUP (ORDER BY job) 
    INTO joblist FROM (SELECT DISTINCT job FROM scott.emp);
  open :r FOR 'select * from (select deptno,job from scott.emp) pivot  
    (count(*) for (job) in ('||joblist||'))';
end;
/
print r
 
DEPTNO                 ANALYST                CLERK                  MANAGER                PRESIDENT              SALESMAN               
---------------------- ---------------------- ---------------------- ---------------------- ---------------------- ---------------------- 
30                     0                      1                      1                      0                      4                      
20                     2                      2                      1                      0                      0                      
10                     0                      1                      1                      1                      0                      
 
 
SELECT job FROM scott.emp WHERE job = 'ANALYST';
JOB       
---------
ANALYST
 
UPDATE scott.emp SET job = 'ANALY,ST' WHERE job = 'ANALYST'
 
SELECT job FROM scott.emp WHERE job = 'ANALYST';
---------
ANALY,ST
 
var r refcursor
declare joblist varchar2(4000);
begin
  SELECT listagg( ''''||job||''' as '||job,',') within GROUP (ORDER BY job) 
    INTO joblist FROM (SELECT DISTINCT job FROM scott.emp);
  open :r FOR 'select * from (select deptno,job from scott.emp) pivot  
    (count(*) for (job) in ('||joblist||'))';
end;
/
print r
 
Error report:
ORA-56901: non-constant expression IS NOT allowed FOR pivot|unpivot VALUES
ORA-06512: at line 5

jko

[justinedr71]

J'ai pas mal de possibilités pour le contenu de ma colonne ccp_libelle.

Globalement, j'ai les caractères de [0..9a..zA..Z], mais aussi [.,:-_<>]...

Êtes-vous sûr que le problème vient du contenu de ma chaine? Comment y remédier?

[jkofr]

Pour les ',' tu peux rajouter ca

Code :

1
2
3
4
5
6
7
8
9
10
11
12

var r refcursor
declare joblist varchar2(4000);
begin
  SELECT listagg( ''''||job||''' as '||replace (job,',',''),',') within GROUP (ORDER BY job) 
    INTO joblist FROM (SELECT DISTINCT  job FROM scott.emp);
  open :r FOR 'select * from (select deptno,job from scott.emp) pivot  
    (count(*) for (job) in ('||joblist||'))';
end;
/
print r

[justinedr71]

J'ai utilisé votre dernière requête.

Voici ma requête de base :

Code sql :

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
 
SELECT *
FROM (SELECT pol_numpol, sor_ident, ver_compteur, ver_libelle, ver_dateoperation, sor_datedebut, sor_datefin, sor_datedeb_fact, sor_datefin_fact, pol_datresil, pol_datechpro, ccp_libelle,
gad_prime_nette, pas_nb_jours_prorata, pri_type_risque, bpp_type, bpp_pp_sexe, bpp_pp_nais_date, bpp_pp_nais_pay_code, mar_libelle, bpp_pp_nb_enfants, bpp_pp_handicape, in_code_etabl,
in_type
FROM f_polices, f_version_police, f_mouvement, f_sit_objet_risque, f_garantie_dyn, f_produitass, f_prd_risque, f_sit_assure, f_p_c_assure, f_p_personne, t_sit_maritale, f_intermediaire,
f_tarif_ass, t_intercalaire, f_tarif_gar, t_cdecpt
WHERE gad_ptrsorid = sor_ident
AND sor_ptrpolid = pol_ident
AND sor_ident = mvt_ptrsuivantid
AND mvt_ptrverid = ver_ident
AND pol_ptrpasid = pas_ident
AND pas_ident = pri_ptrpasident
AND sor_ident = sar_ptrsorid (+)
AND sar_ptrassid = bpass_ident (+)
AND bpass_ptrbppident = bpp_ident (+)
AND bpp_ptrmarid = mar_ident (+)
AND pol_ptrinid = in_ident (+)
AND sor_datetarif = tas_dateffet
AND tas_codeintercalaire = ta_code
AND ta_cle = tga_cletarif
AND tga_code_compta = ccp_code
AND gad_code = tga_code
AND tga_genre = 'GAR'
AND gad_prime_nette IS NOT NULL AND gad_prime_nette != 0
AND (pol_datresil IS NULL OR pol_datresil >= (sysdate - to_yminterval('03-00')))
AND ((sor_datedebut != sor_datefin) OR (sor_datefin IS NULL AND sor_datedebut != pol_datresil) 
OR (sor_datefin IS NULL AND pol_datresil IS NULL AND sor_datedebut != pol_datechpro))
GROUP BY pol_numpol, pol_datresil, pol_datechpro, ver_compteur, ver_libelle, ver_dateoperation, sor_ident, sor_datedebut, sor_datefin, sor_datedeb_fact, sor_datefin_fact, gad_prime_nette, ccp_libelle, pas_nb_jours_prorata, pri_type_risque, bpp_type, bpp_pp_sexe, bpp_pp_nais_date, bpp_pp_nais_pay_code, mar_libelle, bpp_pp_nb_enfants, bpp_pp_handicape, in_code_etabl, in_type )
PIVOT  (SUM(gad_prime_nette) AS prime FOR (ccp_libelle) IN ('Vol' AS Vol, 'Incendie' AS Incendie))
ORDER BY pol_numpol ASC, ver_compteur ASC, sor_ident ASC

Avec pour exemple les deux libellés Vol et Incendie (qui sont parmis les nombreux libellés existants). A noter que je ne connais pas ni le nombre ni le nom des libellés...

Cette requête fonctionne parfaitement, maintenant, la voici concaténée dans votre requête à vous :

Code sql :

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
 
var r refcursor
declare joblist varchar2(4000);
begin
  SELECT listagg( ''''||ccp_libelle||''' as '||REPLACE (ccp_libelle,',',''),',') within GROUP (ORDER BY ccp_libelle) 
    INTO joblist FROM (SELECT DISTINCT ccp_libelle FROM t_cdecpt);
 
      open :r FOR 'SELECT *
FROM (select pol_numpol, sor_ident, ver_compteur, ver_libelle, ver_dateoperation, sor_datedebut, sor_datefin, sor_datedeb_fact, sor_datefin_fact, pol_datresil, pol_datechpro, ccp_libelle,
gad_prime_nette, pas_nb_jours_prorata, pri_type_risque, bpp_type, bpp_pp_sexe, bpp_pp_nais_date, bpp_pp_nais_pay_code, mar_libelle, bpp_pp_nb_enfants, bpp_pp_handicape, in_code_etabl,
in_type
from f_polices, f_version_police, f_mouvement, f_sit_objet_risque, f_garantie_dyn, f_produitass, f_prd_risque, f_sit_assure, f_p_c_assure, f_p_personne, t_sit_maritale, f_intermediaire,
f_tarif_ass, t_intercalaire, f_tarif_gar, t_cdecpt
where gad_ptrsorid = sor_ident
and sor_ptrpolid = pol_ident
and sor_ident = mvt_ptrsuivantid
and mvt_ptrverid = ver_ident
and pol_ptrpasid = pas_ident
and pas_ident = pri_ptrpasident
and sor_ident = sar_ptrsorid (+)
and sar_ptrassid = bpass_ident (+)
and bpass_ptrbppident = bpp_ident (+)
and bpp_ptrmarid = mar_ident (+)
and pol_ptrinid = in_ident (+)
and sor_datetarif = tas_dateffet
and tas_codeintercalaire = ta_code
and ta_cle = tga_cletarif
and tga_code_compta = ccp_code
and gad_code = tga_code
and tga_genre = 'GAR'
and gad_prime_nette is not null and gad_prime_nette != 0
and (pol_datresil is null or pol_datresil >= (sysdate - to_yminterval('03-00')))
and ((sor_datedebut != sor_datefin) or (sor_datefin is null and sor_datedebut != pol_datresil) 
or (sor_datefin is null and pol_datresil is null and sor_datedebut != pol_datechpro))
group by pol_numpol, pol_datresil, pol_datechpro, ver_compteur, ver_libelle, ver_dateoperation, sor_ident, sor_datedebut, sor_datefin, sor_datedeb_fact, sor_datefin_fact, gad_prime_nette, ccp_libelle, pas_nb_jours_prorata, pri_type_risque, bpp_type, bpp_pp_sexe, bpp_pp_nais_date, bpp_pp_nais_pay_code, mar_libelle, bpp_pp_nb_enfants, bpp_pp_handicape, in_code_etabl, in_type ) 
pivot (sum(gad_prime_nette) for (ccp_libelle) in ('||joblist||'))';    
end;
/
print r

Ici elle ne fonctionne pas. Peut-être est-ce du aux ' ' de chaque côté de 'GAR' ou encore '03-00' ou même '||joblist||'...

J'obtiend l'erreur suivante :

ORA-06550: Ligne 28, colonne 18 :
PLS-00103: Symbole "GAR" rencontré à la place d'un des symboles suivants :

* & = - + ; < / > at in is mod remainder not rem
<exposant (**)> <> or != or ~= >= <= <> and or like like2
like4 likec between using || multiset member submultiset

[jkofr]

Et la?

Code :

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
 
 
var r refcursor
declare joblist varchar2(4000);
begin
  SELECT listagg( ''''||ccp_libelle||''' as '||REPLACE (ccp_libelle,',',''),',') within GROUP (ORDER BY ccp_libelle) 
    INTO joblist FROM (SELECT DISTINCT ccp_libelle FROM t_cdecpt);
 
      open :r FOR 'SELECT *
FROM (select pol_numpol, sor_ident, ver_compteur, ver_libelle, ver_dateoperation, sor_datedebut, sor_datefin, sor_datedeb_fact, sor_datefin_fact, pol_datresil, pol_datechpro, ccp_libelle,
gad_prime_nette, pas_nb_jours_prorata, pri_type_risque, bpp_type, bpp_pp_sexe, bpp_pp_nais_date, bpp_pp_nais_pay_code, mar_libelle, bpp_pp_nb_enfants, bpp_pp_handicape, in_code_etabl,
in_type
from f_polices, f_version_police, f_mouvement, f_sit_objet_risque, f_garantie_dyn, f_produitass, f_prd_risque, f_sit_assure, f_p_c_assure, f_p_personne, t_sit_maritale, f_intermediaire,
f_tarif_ass, t_intercalaire, f_tarif_gar, t_cdecpt
where gad_ptrsorid = sor_ident
and sor_ptrpolid = pol_ident
and sor_ident = mvt_ptrsuivantid
and mvt_ptrverid = ver_ident
and pol_ptrpasid = pas_ident
and pas_ident = pri_ptrpasident
and sor_ident = sar_ptrsorid (+)
and sar_ptrassid = bpass_ident (+)
and bpass_ptrbppident = bpp_ident (+)
and bpp_ptrmarid = mar_ident (+)
and pol_ptrinid = in_ident (+)
and sor_datetarif = tas_dateffet
and tas_codeintercalaire = ta_code
and ta_cle = tga_cletarif
and tga_code_compta = ccp_code
and gad_code = tga_code
and tga_genre = ''GAR''
and gad_prime_nette is not null and gad_prime_nette != 0
and (pol_datresil is null or pol_datresil >= (sysdate - to_yminterval(''03-00'')))
and ((sor_datedebut != sor_datefin) or (sor_datefin is null and sor_datedebut != pol_datresil) 
or (sor_datefin is null and pol_datresil is null and sor_datedebut != pol_datechpro))
group by pol_numpol, pol_datresil, pol_datechpro, ver_compteur, ver_libelle, ver_dateoperation, sor_ident, sor_datedebut, sor_datefin, sor_datedeb_fact, sor_datefin_fact, gad_prime_nette, ccp_libelle, pas_nb_jours_prorata, pri_type_risque, bpp_type, bpp_pp_sexe, bpp_pp_nais_date, bpp_pp_nais_pay_code, mar_libelle, bpp_pp_nb_enfants, bpp_pp_handicape, in_code_etabl, in_type ) 
pivot (sum(gad_prime_nette) for (ccp_libelle) in ('||joblist||'))';    
end;
/
print r

[justinedr71]

ORA-00917: virgule absente
ORA-06512: à ligne 6

... Il recommence avec cette histoire de virgule...

[justinedr71]

Tu as bien mis deux quotes (') à la suite, ou bien ce sont des guillemets (")?

[jkofr]

Code :

1
2
3
4
and tga_genre = ''GAR''
and gad_prime_nette is not null and gad_prime_nette != 0
and (pol_datresil is null or pol_datresil >= (sysdate - to_yminterval(''03-00'')))

[justinedr71]

Ok! C'est bien ce que j'ai...
Pourquoi ce problème de virgule revient...?

[jkofr]

Tiens, essaie ca!

En fait le replace est inutile!

Code :

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
 
 
var r refcursor
declare joblist varchar2(4000);
begin
  SELECT listagg( ''''||ccp_libelle||''' as "'||ccp_libelle||'"',',') within GROUP (ORDER BY ccp_libelle) 
    INTO joblist FROM (SELECT DISTINCT ccp_libelle FROM t_cdecpt);
 
      open :r FOR 'SELECT *
FROM (select pol_numpol, sor_ident, ver_compteur, ver_libelle, ver_dateoperation, sor_datedebut, sor_datefin, sor_datedeb_fact, sor_datefin_fact, pol_datresil, pol_datechpro, ccp_libelle,
gad_prime_nette, pas_nb_jours_prorata, pri_type_risque, bpp_type, bpp_pp_sexe, bpp_pp_nais_date, bpp_pp_nais_pay_code, mar_libelle, bpp_pp_nb_enfants, bpp_pp_handicape, in_code_etabl,
in_type
from f_polices, f_version_police, f_mouvement, f_sit_objet_risque, f_garantie_dyn, f_produitass, f_prd_risque, f_sit_assure, f_p_c_assure, f_p_personne, t_sit_maritale, f_intermediaire,
f_tarif_ass, t_intercalaire, f_tarif_gar, t_cdecpt
where gad_ptrsorid = sor_ident
and sor_ptrpolid = pol_ident
and sor_ident = mvt_ptrsuivantid
and mvt_ptrverid = ver_ident
and pol_ptrpasid = pas_ident
and pas_ident = pri_ptrpasident
and sor_ident = sar_ptrsorid (+)
and sar_ptrassid = bpass_ident (+)
and bpass_ptrbppident = bpp_ident (+)
and bpp_ptrmarid = mar_ident (+)
and pol_ptrinid = in_ident (+)
and sor_datetarif = tas_dateffet
and tas_codeintercalaire = ta_code
and ta_cle = tga_cletarif
and tga_code_compta = ccp_code
and gad_code = tga_code
and tga_genre = ''GAR''
and gad_prime_nette is not null and gad_prime_nette != 0
and (pol_datresil is null or pol_datresil >= (sysdate - to_yminterval(''03-00'')))
and ((sor_datedebut != sor_datefin) or (sor_datefin is null and sor_datedebut != pol_datresil) 
or (sor_datefin is null and pol_datresil is null and sor_datedebut != pol_datechpro))
group by pol_numpol, pol_datresil, pol_datechpro, ver_compteur, ver_libelle, ver_dateoperation, sor_ident, sor_datedebut, sor_datefin, sor_datedeb_fact, sor_datefin_fact, gad_prime_nette, ccp_libelle, pas_nb_jours_prorata, pri_type_risque, bpp_type, bpp_pp_sexe, bpp_pp_nais_date, bpp_pp_nais_pay_code, mar_libelle, bpp_pp_nb_enfants, bpp_pp_handicape, in_code_etabl, in_type ) 
pivot (sum(gad_prime_nette) for (ccp_libelle) in ('||joblist||'))';    
end;
/
print r

jko

[justinedr71]

Ah! On change d'erreur :

ORA-00972: l'identificateur est trop long
ORA-06512: à ligne 6

Et il me surligne la ligne :
declare joblist varchar2(4000);

...

[jkofr]

Bien,

Voici un exemple pour que tu comprenne

Code :

1
2
3
4
5
6
7
 
SELECT 1 AS MonLibelledecolonnepourtesterlalongeurmax FROM dual;
 
SQL Error: ORA-00972: identifier IS too long
00972. 00000 -  "identifier is too long"
*Cause:    An identifier WITH more than 30 characters was specified.
*Action:   Specify at most 30 characters.

En bref, ton libelle de colonne ne peut pas faire plus de 30 caractères.

Soit tu fais un substr (ccp_libelle,1,30), soit tu n'as pas de solution facile...

Bon courage
jko

[justinedr71]

Super!
Merci beaucoup!!!

Voici ma requete finale, qui fonctionne :

Code sql :

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
 
var r refcursor
declare joblist varchar2(4000);
begin
  SELECT listagg( ''''||libelle||''' as "'||libelle||'"',',') within GROUP (ORDER BY libelle) 
    INTO joblist FROM (SELECT DISTINCT substr(ccp_libelle,1,30) AS libelle FROM t_cdecpt);
 
      open :r FOR 'SELECT *
FROM (select pol_numpol, sor_ident, ver_compteur, ver_libelle, ver_dateoperation, sor_datedebut, sor_datefin, sor_datedeb_fact, sor_datefin_fact, pol_datresil, pol_datechpro, ccp_libelle,
gad_prime_nette, pas_nb_jours_prorata, pri_type_risque, bpp_type, bpp_pp_sexe, bpp_pp_nais_date, bpp_pp_nais_pay_code, mar_libelle, bpp_pp_nb_enfants, bpp_pp_handicape, in_code_etabl,
in_type
from f_polices, f_version_police, f_mouvement, f_sit_objet_risque, f_garantie_dyn, f_produitass, f_prd_risque, f_sit_assure, f_p_c_assure, f_p_personne, t_sit_maritale, f_intermediaire,
f_tarif_ass, t_intercalaire, f_tarif_gar, t_cdecpt
where gad_ptrsorid = sor_ident
and sor_ptrpolid = pol_ident
and sor_ident = mvt_ptrsuivantid
and mvt_ptrverid = ver_ident
and pol_ptrpasid = pas_ident
and pas_ident = pri_ptrpasident
and sor_ident = sar_ptrsorid (+)
and sar_ptrassid = bpass_ident (+)
and bpass_ptrbppident = bpp_ident (+)
and bpp_ptrmarid = mar_ident (+)
and pol_ptrinid = in_ident (+)
and sor_datetarif = tas_dateffet
and tas_codeintercalaire = ta_code
and ta_cle = tga_cletarif
and tga_code_compta = ccp_code
and gad_code = tga_code
and tga_genre = ''GAR''
and gad_prime_nette is not null and gad_prime_nette != 0
and (pol_datresil is null or pol_datresil >= (sysdate - to_yminterval(''03-00'')))
and ((sor_datedebut != sor_datefin) or (sor_datefin is null and sor_datedebut != pol_datresil) 
or (sor_datefin is null and pol_datresil is null and sor_datedebut != pol_datechpro))
group by pol_numpol, pol_datresil, pol_datechpro, ver_compteur, ver_libelle, ver_dateoperation, sor_ident, sor_datedebut, sor_datefin, sor_datedeb_fact, sor_datefin_fact, gad_prime_nette, ccp_libelle, pas_nb_jours_prorata, pri_type_risque, bpp_type, bpp_pp_sexe, bpp_pp_nais_date, bpp_pp_nais_pay_code, mar_libelle, bpp_pp_nb_enfants, bpp_pp_handicape, in_code_etabl, in_type ) 
pivot (sum(gad_prime_nette) for (ccp_libelle) in ('||joblist||'))';    
end;
/
print r

[justinedr71]

Ah non une dernière chose. Comment puis-je faire pour que même les garanties ayant un libelle > 30 soient prises en compte.

Ex, j'ai une garantie avec un libelle 'Protection juiridique automobile', 2 caractères de trop!!! Seulement, il me met bien le champ 'Protection juiridique automobi' mais du coup il ne prend pas en compte les valeurs des primes associées à cette garantie dans ma table, ce que je pense normale, car dans ma table ces primes sont associées à 'Protection juiridique automobile' et non 'Protection juiridique automobi'... Comment faire pour palier à cela.

Je pensais aux expressions régulières avec '*' mais je ne sais pas comment le formuler...

Merci pour vos réponses