Re: [TiLP-devel] 86 file format compatibility patch

[<ti...@li...>]

Hi Adrian,

thank you for your patch ! Given that I will not be able to test it (I=20
don't have TI85/86 anymore), I would like to know whether you have=20
tested it ?

If yes, I will apply it. Else, I will have to take a look more thourougly=
...

Thanks, Romain.

Adrian Mettler a =E9crit :
> This is sent from my primary (more or less spam-free... knock on wood)=20
> address.
>=20
> Sorry about the mailer splitting some of the lines.
> --Adrian
>=20
>=20
> --- files8x.c.bak       2003-08-27 02:30:44.000000000 -0700
> +++ files8x.c   2003-11-27 10:38:33.000000000 -0800
> @@ -194,6 +194,8 @@
>    int i;
>    int ti83p_flag =3D 0;
>    uint8_t name_length =3D 8;     // ti85/86 only
> +  uint16_t data_size;
> +  uint8_t test_space;
>    char signature[9];
>=20
>    if (!tifiles_is_a_ti_file(filename))
> @@ -213,18 +215,45 @@
>      return ERR_INVALID_FILE;
>    fskip(f, 3);
>    fread_n_chars(f, 42, content->comment);
> -  fread_word(f, NULL);
> +  fread_word(f, &data_size);
>=20
>    // search for the number of entries by parsing the whole file
>    offset =3D ftell(f);
>    for (i =3D 0;; i++) {
> +    int current_offset =3D (int)ftell(f);
> +    /* We are done finding entries once we reach the end of the data=20
> segment
> +     * as defined in the header.  This works better than magic numbers=
, as
> +     * as there exist files in the wild with incorrect magic numbers t=
hat
> +     * transmit correctly with TI's software and with this code.
> +     */
> +    //printf("current_offset: %x, offset + data_size: %x\n",
> +    //    current_offset, (uint16_t) offset + data_size);
> +    if (current_offset >=3D offset + data_size)
> +      break;
> +
>      fread_word(f, &tmp);
>      if (tmp =3D=3D 0x0D)
>        ti83p_flag =3D !0;         // true TI83+ file (2 extra bytes)
> -    if ((tmp !=3D 0x0B) && (tmp !=3D 0x0C) && (tmp !=3D 0x0D))
> -      break;
> +    //if ((tmp !=3D 0x0B) && (tmp !=3D 0x0C) && (tmp !=3D 0x0D))
> +    //  break;
>      if (is_ti8586(content->calc_type))
> -      fskip(f, 12);
> +    {
> +      /* name may follow one of three conventions: padded with zeroes
> +       * (most correct, generated by TI's software), padded with null=20
> bytes,
> +       * or unpadded.  TI's software accepts all three, so we should t=
oo.
> +       */
> +      fskip(f, 3);
> +      fread_byte(f, &name_length);
> +      fskip(f, name_length);
> +
> +      fread_byte(f, &test_space);
> +      fskip (f, -1);
> +
> +      //printf("test_space: %x at %x\n", test_space, (int)ftell(f));
> +
> +      if (test_space =3D=3D '\0' || test_space =3D=3D ' ')  // skip pa=
dding if=20
> padded
> +       fskip(f, 8 - name_length);
> +    }
>      else if (content->calc_type =3D=3D CALC_TI83P)
>        fskip(f, 13);
>      else
> @@ -254,7 +283,16 @@
>      tixx_translate_varname(entry->name, entry->trans,
>                            entry->type, content->calc_type);
>      if (is_ti8586(content->calc_type))
> -      fskip(f, 8 - name_length);
> +    {
> +      /* Again, compatibility with padded and unpadded files
> +       */
> +      fread_byte(f, &test_space);
> +      fskip (f, -1);
> +
> +      if (test_space =3D=3D '\0' || test_space =3D=3D ' ')  // skip pa=
dding if=20
> padded
> +       fskip(f, 8 - name_length);
> +    }
> +
>      if (ti83p_flag) {
>        entry->attr =3D (fgetc(f) =3D=3D 0x80 ? ATTRB_ARCHIVED : ATTRB_N=
ONE);
>        fgetc(f);
> @@ -639,7 +677,7 @@
>      fwrite_word(f, content->data_length4);
>      fwrite(content->data_part4, 1, content->data_length4, f);
>    }
> -  // checksum =3D sum of all bytes in bachup headers and data num_entr=
ies
> +  // checksum =3D sum of all bytes in backup headers and data num_entr=
ies
>    sum =3D 0;
>    sum +=3D 9;
>    sum +=3D
>=20
>=20
>=20
>=20
> --- Romain Li=E9vin <ro...@us...> wrote:
>  >
>  > Hi,
>  >
>  >
>  >
>  > could you send to me your patch in unified diff
>  > format (diff
>  >
>  > -Naur) ?
>  >
>  >
>  >
>  > You can mail me at <ro...@ti...> or
>  > til...@li....
>  >
>  >
>  >
>  > It will be easier to communicate by this way rather
>  > than by
>  >
>  > the BugTracker.
>  >
>  >
>  >
>  > thx, roms.
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by: IBM Linux Tutorials
> Free Linux tutorial presented by Daniel Robbins, President and CEO of
> GenToo technologies. Learn everything from fundamentals to system
> administration.http://ads.osdn.com/?ad_id=1470&alloc_id638&op=3Dclick
> _______________________________________________
> TiLP-devel mailing list
> TiL...@li...
> https://lists.sourceforge.net/lists/listinfo/tilp-devel
>=20
>=20


--=20
Romain Li=E9vin (roms):         <ro...@ti...>
Web site:                     http://www.lievin.net
"Linux, y'a moins bien mais c'est plus cher !"