The problem with MEMO

[Auge_Ohr]

Eugene we try to help you as much as possible but to revise your code to look for FUNCTION F4_2_2_3() i have no time

my Tip to create a small App to check your FUNCTION F4_2_2_3() is for you !
if you have made a small App than you can upload it and ask other User to check it if they have time.

---

about Array : as i say you can store hole Array binarie to Memo Type "V" like in Project "Delone Triangulation"

Code: Select all

   REPLACE TRIMEMO->POINTS    WITH Var2Bin(aPoints)
   REPLACE TRIMEMO->TRIANGLE  WITH Var2Bin(aTriangles)

and back from Memo

Code: Select all

   ::points    := Bin2Var(TRIMEMO->POINTS  )
   ::Triangles := Bin2Var(TRIMEMO->TRIANGLE)

as i use same DBE Settings like Express++ default (?)

Code: Select all

FUNCTION DC_LoadRdds
...
  DbeInfo( COMPONENT_DATA, FOXDBE_CREATE_2X, .T. )
  DbeInfo( COMPONENT_DATA, FOXDBE_LOCKMODE , FOXDBE_LOCKMODE_2X )
  DbeInfo( COMPONENT_ORDER, CDXDBE_MODE , CDXDBE_FOXPRO2X )

that is FoxPro 2.x compatible so this Code to create Database/Memo must work with Express++ default

Code: Select all

FUNCTION CRE_TRIPLEMEMO(datei,alias,id)
  LOCAL p,field_list:={}
  IF VALTYPE(datei)!="C"
    datei := "TRIMEMO.DBF"
  ENDIF
  IF VALTYPE(alias)!="C"
    p := AT(".",datei)
    alias := IF(p>0,SUBSTR(datei,1,p-1),datei)
  ENDIF
  IF VALTYPE(id)!="N"
    id := 0
  ENDIF
  SELECT (id)
  IF !FILE(datei)
    AADD(field_list,{"NUM"     ,"N",  5,0})
    AADD(field_list,{"DESC"    ,"C",100,0})
    AADD(field_list,{"POINTS"  ,"V",  8,0})
    AADD(field_list,{"RIBS"    ,"V",  8,0})
    AADD(field_list,{"TRIANGLE","V",  8,0})
    DBCREATE(datei,field_list,"FOXCDX")
  ENDIF
RETURN(.t.)

[Eugene Lutsenko]

Hello, Auge_Ohr!

I just thought you might be interested. Maybe you also use tree clustering somewhere in Your home. Sorry. I am very grateful to you for your help. I don't have anyone else to talk to but you. You help me a lot. I think I'm going to solve this problem with the memoir by simply abandoning it. I myself have no problem realizing that I wanted to do with memo. In the meantime, everything works well with simple text fields.

[Victorio]

Eugene,
Maybe this is off topic, but can you use several C fields instead of memo field ?
Memo files in our office often disturbed, often after power failure or failure server disks.

I use to store text in dbf in several fields C 255 and in program link it to one string. but with index can be problem because limit of length index key.
Better way will be use sql database...

[Eugene Lutsenko]

Victorio!

Combining "C" several fields programmatically is not a bad idea. And simply realized

[Victorio]

only for example source for read text file to dbf :

nHandle:=DC_TxtOpen(infuvi)

* cyklus pre načítanie riadkov súboru
do while !DC_TxtEof(nHandle)
* načítanie riadku z txt súboru
cLine:=DC_TxtLine(nHandle)
* pridanie prázdneho riadku do databázy
IF 1->(DC_AddRec(5))
* zápis riadku do databázy

* kontrola na dĺžku riadku ak dlhší, rozdelí do dvoch
if len(cLine)<=254
replace riadok1 with cLine

elseif len(cLine)>1*254 .and. len(cLine)<=2*254
replace riadok1 with substr(cLine,1,254)
replace riadok2 with substr(cLine,(1*254)+1,254)

elseif len(cLine)>2*254 .and. len(cLine)<=3*254
replace riadok1 with substr(cLine,1,254)
replace riadok2 with substr(cLine,(2*254)+1,254)
replace riadok3 with substr(cLine,(3*254)+1,254)

elseif len(cLine)>3*254 .and. len(cLine)<=4*254
replace riadok1 with substr(cLine,1,254)
replace riadok2 with substr(cLine,(1*254)+1,254)
replace riadok3 with substr(cLine,(2*254)+1,254)
replace riadok4 with substr(cLine,(3*254)+1,254)

elseif len(cLine)>4*254 .and. len(cLine)<=5*254
replace riadok1 with substr(cLine,1,254)
replace riadok2 with substr(cLine,(1*254)+1,254)
replace riadok3 with substr(cLine,(2*254)+1,254)
replace riadok4 with substr(cLine,(3*254)+1,254)
replace riadok5 with substr(cLine,(4*254)+1,254)

endif
* odomknutie záznamu
1->(dbRUnlock())
ENDIF
DC_TxtSkip(nHandle,1)
enddo
* zatvorenie text.súboru
DC_TxtClose(nHandle)

[Eugene Lutsenko]

Thanks, Victorio!

Now the memo fields are not needed at all, especially since they still do not work

[Eugene Lutsenko]

Thanks to your fresh ideas, I solved this issue in an elementary and radical way.

Instead of REPLACE use StrFile. Instead of the field value, I use the value of the variable to which the text and the file are assigned. Then I delete temporary TXT files. Maybe not very elegant. But it works well and is not difficult to alter the program

Code: Select all

           REPLACE NameCls_Fu  WITH mNameClustFu

           StrFile(ALLTRIM(mNameClustFu),  M_PathAppl+"\ClsClustTree\NameClsF"+"-"+STRTRAN(STR(mNumMod,2)," ","0")+"-"+STRTRAN(STR(KODCL_NEW,5)," ","0")+".txt")     // Запись     текстового файла NameClsF-##-#####.txt, где ##-номер модели, #####-KODCL_NEW

*          mNameClustFu = FileStr(M_PathAppl+"\ClsClustTree\NameClsF"+"-"+STRTRAN(STR(mNumMod,2)," ","0")+"-"+STRTRAN(STR(KODCL_NEW,5)," ","0")+".txt")              // Считывание текстового файла NameClsF-##-#####.txt, где ##-номер модели, #####-KODCL_NEW

[/size]

Note the cluster nesting formula at the bottom (indicated by an arrow). Now there are no restrictions on the length of this formula. So I can draw up to 100 clusters on a graphical form. If you use just text boxes, you can handle no more than 50 clusters correctly