SQLite: Check-in [7c15d762]

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Overview

Comment:	The ".import" command of the shell, and the csv virtual table extension both ignore a single UTF-8 BOM at the beginning of their input.
Downloads:	Tarball \| ZIP archive \| SQL archive
Timelines:	family \| ancestors \| descendants \| both \| trunk
Files:	files \| file ages \| folders
SHA3-256:	7c15d762d99c2e3e534cd35dfe25ddcd317637eb1f2655fd24c2dd5f9d5a7613
User & Date:	drh 2017-06-26 18:42:23

Context

2017-06-26
21:08		Add the -withoutnulls option to the "db eval" method in the TCL interface. (check-in: 18f0616e user: drh tags: trunk)
18:42		The ".import" command of the shell, and the csv virtual table extension both ignore a single UTF-8 BOM at the beginning of their input. (check-in: 7c15d762 user: drh tags: trunk)
2017-06-24
19:21		Disable shell tests for the .schema command if virtual tables are not available. (check-in: c8186874 user: drh tags: trunk)

Changes

Hide Diffs Unified Diffs Ignore Whitespace Patch

Changes to ext/misc/csv.c.

typedef struct CsvReader CsvReader;
struct CsvReader {
  FILE *in;              /* Read the CSV text from this input stream */
  char *z;               /* Accumulated text for a field */
  int n;                 /* Number of bytes in z */
  int nAlloc;            /* Space allocated for z[] */
  int nLine;             /* Current line number */

  char cTerm;            /* Character that terminated the most recent field */
  size_t iIn;            /* Next unread character in the input buffer */
  size_t nIn;            /* Number of characters in the input buffer */
  char *zIn;             /* The input buffer */
  char zErr[CSV_MXERR];  /* Error message */
};

................................................................................
/* Initialize a CsvReader object */
static void csv_reader_init(CsvReader *p){
  p->in = 0;
  p->z = 0;
  p->n = 0;
  p->nAlloc = 0;
  p->nLine = 0;

  p->nIn = 0;
  p->zIn = 0;
  p->zErr[0] = 0;
}

/* Close and reset a CsvReader object */
static void csv_reader_reset(CsvReader *p){
................................................................................
        }
      }
      if( csv_append(p, (char)c) ) return 0;
      ppc = pc;
      pc = c;
    }
  }else{















    while( c>',' || (c!=EOF && c!=',' && c!='\n') ){
      if( csv_append(p, (char)c) ) return 0;
      c = csv_getc(p);
    }
    if( c=='\n' ){
      p->nLine++;
      if( p->n>0 && p->z[p->n-1]=='\r' ) p->n--;
    }
    p->cTerm = (char)c;
  }
  if( p->z ) p->z[p->n] = 0;

  return p->z;
}


/* Forward references to the various virtual table methods implemented
** in this file. */
static int csvtabCreate(sqlite3*, void*, int, const char*const*,








>







 







>







 







>
>
>
>
>
>
>
>
>
>
>
>
>
>
>











>

typedef struct CsvReader CsvReader;
struct CsvReader {
  FILE *in;              /* Read the CSV text from this input stream */
  char *z;               /* Accumulated text for a field */
  int n;                 /* Number of bytes in z */
  int nAlloc;            /* Space allocated for z[] */
  int nLine;             /* Current line number */
  int bNotFirst;         /* True if prior text has been seen */
  char cTerm;            /* Character that terminated the most recent field */
  size_t iIn;            /* Next unread character in the input buffer */
  size_t nIn;            /* Number of characters in the input buffer */
  char *zIn;             /* The input buffer */
  char zErr[CSV_MXERR];  /* Error message */
};

................................................................................
/* Initialize a CsvReader object */
static void csv_reader_init(CsvReader *p){
  p->in = 0;
  p->z = 0;
  p->n = 0;
  p->nAlloc = 0;
  p->nLine = 0;
  p->bNotFirst = 0;
  p->nIn = 0;
  p->zIn = 0;
  p->zErr[0] = 0;
}

/* Close and reset a CsvReader object */
static void csv_reader_reset(CsvReader *p){
................................................................................
        }
      }
      if( csv_append(p, (char)c) ) return 0;
      ppc = pc;
      pc = c;
    }
  }else{
    /* If this is the first field being parsed and it begins with the
    ** UTF-8 BOM  (0xEF BB BF) then skip the BOM */
    if( (c&0xff)==0xef && p->bNotFirst==0 ){
      csv_append(p, c);
      c = csv_getc(p);
      if( (c&0xff)==0xbb ){
        csv_append(p, c);
        c = csv_getc(p);
        if( (c&0xff)==0xbf ){
          p->bNotFirst = 1;
          p->n = 0;
          return csv_read_one_field(p);
        }
      }
    }
    while( c>',' || (c!=EOF && c!=',' && c!='\n') ){
      if( csv_append(p, (char)c) ) return 0;
      c = csv_getc(p);
    }
    if( c=='\n' ){
      p->nLine++;
      if( p->n>0 && p->z[p->n-1]=='\r' ) p->n--;
    }
    p->cTerm = (char)c;
  }
  if( p->z ) p->z[p->n] = 0;
  p->bNotFirst = 1;
  return p->z;
}


/* Forward references to the various virtual table methods implemented
** in this file. */
static int csvtabCreate(sqlite3*, void*, int, const char*const*,

Changes to src/shell.c.

struct ImportCtx {
  const char *zFile;  /* Name of the input file */
  FILE *in;           /* Read the CSV text from this input stream */
  char *z;            /* Accumulated text for a field */
  int n;              /* Number of bytes in z */
  int nAlloc;         /* Space allocated for z[] */
  int nLine;          /* Current line number */

  int cTerm;          /* Character that terminated the most recent field */
  int cColSep;        /* The column separator character.  (Usually ",") */
  int cRowSep;        /* The row separator character.  (Usually "\n") */
};

/* Append a single byte to z[] */
static void import_append_char(ImportCtx *p, int c){
................................................................................
        break;
      }
      import_append_char(p, c);
      ppc = pc;
      pc = c;
    }
  }else{















    while( c!=EOF && c!=cSep && c!=rSep ){
      import_append_char(p, c);
      c = fgetc(p->in);
    }
    if( c==rSep ){
      p->nLine++;
      if( p->n>0 && p->z[p->n-1]=='\r' ) p->n--;
    }
    p->cTerm = c;
  }
  if( p->z ) p->z[p->n] = 0;

  return p->z;
}

/* Read a single field of ASCII delimited text.
**
**   +  Input comes from p->in.
**   +  Store results in p->z of length p->n.  Space to hold p->z comes








>







 







>
>
>
>
>
>
>
>
>
>
>
>
>
>
>











>

struct ImportCtx {
  const char *zFile;  /* Name of the input file */
  FILE *in;           /* Read the CSV text from this input stream */
  char *z;            /* Accumulated text for a field */
  int n;              /* Number of bytes in z */
  int nAlloc;         /* Space allocated for z[] */
  int nLine;          /* Current line number */
  int bNotFirst;      /* True if one or more bytes already read */
  int cTerm;          /* Character that terminated the most recent field */
  int cColSep;        /* The column separator character.  (Usually ",") */
  int cRowSep;        /* The row separator character.  (Usually "\n") */
};

/* Append a single byte to z[] */
static void import_append_char(ImportCtx *p, int c){
................................................................................
        break;
      }
      import_append_char(p, c);
      ppc = pc;
      pc = c;
    }
  }else{
    /* If this is the first field being parsed and it begins with the
    ** UTF-8 BOM  (0xEF BB BF) then skip the BOM */
    if( (c&0xff)==0xef && p->bNotFirst==0 ){
      import_append_char(p, c);
      c = fgetc(p->in);
      if( (c&0xff)==0xbb ){
        import_append_char(p, c);
        c = fgetc(p->in);
        if( (c&0xff)==0xbf ){
          p->bNotFirst = 1;
          p->n = 0;
          return csv_read_one_field(p);
        }
      }
    }
    while( c!=EOF && c!=cSep && c!=rSep ){
      import_append_char(p, c);
      c = fgetc(p->in);
    }
    if( c==rSep ){
      p->nLine++;
      if( p->n>0 && p->z[p->n-1]=='\r' ) p->n--;
    }
    p->cTerm = c;
  }
  if( p->z ) p->z[p->n] = 0;
  p->bNotFirst = 1;
  return p->z;
}

/* Read a single field of ASCII delimited text.
**
**   +  Input comes from p->in.
**   +  Store results in p->z of length p->n.  Space to hold p->z comes

Changes to test/shell5.test.

  set res [catchcmd "test.db" {.import shell5.csv t1
SELECT COUNT(*) FROM t1;}]
} {0 7}

do_test shell5-1.4.10.2 {
  catchcmd "test.db" {SELECT b FROM t1 WHERE a='7';}
} {0 {Now is the time for all good men to come to the aid of their country.}}































# check importing very long field
do_test shell5-1.5.1 {
  set str [string repeat X 999]
  set in [open shell5.csv w]
  puts $in "8|$str"
  close $in
................................................................................
  for {set i 1} {$i<$cols} {incr i} {
    append data "$i|"
  }
  append data "$cols"
  set in [open shell5.csv w]
  puts $in $data
  close $in

  set res [catchcmd "test.db" {.import shell5.csv t2
SELECT COUNT(*) FROM t2;}]
} {0 1}

# try importing a large number of rows
set rows 9999
do_test shell5-1.7.1 {
  set in [open shell5.csv w]








>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>







 







>
|

  set res [catchcmd "test.db" {.import shell5.csv t1
SELECT COUNT(*) FROM t1;}]
} {0 7}

do_test shell5-1.4.10.2 {
  catchcmd "test.db" {SELECT b FROM t1 WHERE a='7';}
} {0 {Now is the time for all good men to come to the aid of their country.}}

# import file with 2 rows, 2 columns and an initial BOM
#
do_test shell5-1.4.11 {
  set in [open shell5.csv wb]
  puts $in "\xef\xbb\xbf2|3"
  puts $in "4|5"
  close $in
  set res [catchcmd "test.db" {CREATE TABLE t2(x INT, y INT);
.import shell5.csv t2
.mode quote
.header on
SELECT * FROM t2;}]
 string map {\n | \n\r |} $res
} {0 {'x','y'|2,3|4,5}}

# import file with 2 rows, 2 columns or text with an initial BOM
#
do_test shell5-1.4.12 {
  set in [open shell5.csv wb]
  puts $in "\xef\xbb\xbf\"two\"|3"
  puts $in "4|5"
  close $in
  set res [catchcmd "test.db" {DELETE FROM t2;
.import shell5.csv t2
.mode quote
.header on
SELECT * FROM t2;}]
 string map {\n | \n\r |} $res
} {0 {'x','y'|'two',3|4,5}}

# check importing very long field
do_test shell5-1.5.1 {
  set str [string repeat X 999]
  set in [open shell5.csv w]
  puts $in "8|$str"
  close $in
................................................................................
  for {set i 1} {$i<$cols} {incr i} {
    append data "$i|"
  }
  append data "$cols"
  set in [open shell5.csv w]
  puts $in $data
  close $in
  set res [catchcmd "test.db" {DROP TABLE IF EXISTS t2;
.import shell5.csv t2
SELECT COUNT(*) FROM t2;}]
} {0 1}

# try importing a large number of rows
set rows 9999
do_test shell5-1.7.1 {
  set in [open shell5.csv w]