imprinter - TWAIN Condition Code: TWAIN triplet is out of expected sequence

[dmccallie]

Hello again,

Trying to get an imprinter to work on an older Fujitsu 5900C scanner. Using a very old SANE driver, this scanner's imprinter works just fine, so I know it's supported by the hardware.

Here are the *_PRINTER capabilities listed in the JSON output of the demo32 program:

      "imprinter-supported": false,
 ..... 
          "name": "CAP_PRINTER",
          "value": 4134,
          "type": "standard"
        },
        {
          "name": "CAP_PRINTERENABLED",
          "value": 4135,
          "type": "standard"
        },
        {
          "name": "CAP_PRINTERINDEX",
          "value": 4136,
          "type": "standard"
        },
        {
          "name": "CAP_PRINTERMODE",
          "value": 4137,
          "type": "standard"
        },
        {
          "name": "CAP_PRINTERSTRING",
          "value": 4138,
          "type": "standard"
        },
        {
          "name": "CAP_PRINTERSUFFIX",
          "value": 4139,
          "type": "standard"
        },

In my python code, I was able to successfully call DTWAIN_EnumTwainPrinters, DTWAIN_ArrayGetCount (returns 1), and then DTWAIN_ArrayGetAtLong (returns 0 for the 0-the element) There is only one imprinter so I assume '0' makes sense?

After that, for some reason DTWAIN_SetPrinter(source, 0, True) as well as DTWAIN_SetPrinter(source, 0, False) both fail (call returns False) with last error message = 0

However, a subsequent DTWAIN_GetPrinter successfully returns 0 as the printer number then DTWAIN_EnablePrinter returns successfully.

Then I try to call DTWAIN_SetPrinterStartNumber during the callbacks, passing in the "image number" that should be imprinted on each scan. I call this the first time in response to DTWAIN_TN_ACQUIRESTARTED and then subsequently in response to DTWAIN_TN_PAGECONTINUE.

In both these cases DTWAIN_SetPrinterStartNumber() fails with the error Code -1109 - TWAIN Condition Code: TWAIN triplet is out of expected sequence

Needless to say, nothing prints (though scanning works fine.)

Suggestions?

[dynarithmic]

1. It's strange that "imprinter-supported" appears as false and not true. The Avision 176U I have shows "imprinter-supported": true, along with the CAP_PRINTER capability being supported.

2. The imprinter type of 0 is a "top before" imprinter. You can get the actual TWAIN specification constant name of the printer by calling DTWAIN_GetTwainNameFromConstant(DTWAIN_CONSTANT_TWPR, 0, input_string, input_string_length), and on return input_string will contain TWPR_IMPRINTERTOPBEFORE. The call consists of the constant category (you will need to know the constant's category, in this case, TWPR) for the first argument. The second argument is the 0 you got when you queried the array elements.

3. The DTWAIN_SetPrinterStartNumber issue -- unfortunately I cannot really debug this, as I don't have the device, let alone the driver. The most I can tell you is to experiment with setting the capability value CAP_PRINTERINDEX, as this is basically all that DTWAIN is doing when you call DTWAIN_SetPrinterStartNumber. The Avision scanner doesn't support setting the start number, so I could not test this.

You may also try DTWAIN_SetCapValuesEx2 and use DTWAIN_CONTONEVALUE as the container to use when setting CAP_PRINTERINDEX. For example:

DTWAIN_ARRAY pArray;
// assume pArray is filled with the start number...
DTWAIN_SetCapValuesEx2(Source, CAP_PRINTERINDEX, DTWAIN_CAPSET, DTWAIN_CONTONEVALUE, DTWAIN_DEFAULT, pArray);

Since CAP_PRINTERINDEX can have a container of both DTWAIN_CONTRANGE and DTWAIN_CONTONEVALUE, then explicitly having DTWAIN_CONTONEVALUE may have an effect (but it shouldn't, since DTWAIN tries both combinations).

[dmccallie]

Thanks.

I realize you can't test or inspect driver code, but do you think that setting a printer string to would work, based on the CAP_PRINTERSTRING capacity being listed? We'd actually like to format the image number into a standard string with a specific format, so that's actually a better solution than just setting the printer index. Something like "00n-00m"

Does calling these imprinter messages from within the Callback make sense? Will that be the right Twain "state?"

Does that error message "Code -1109 - TWAIN Condition Code: TWAIN triplet is out of expected sequence" give us any clue?

[dynarithmic]

Does calling these imprinter messages from within the Callback make sense? Will that be the right Twain "state?"

Try setting the capabilities directly after opening the Source (state 4). The DTWAIN_TN_ACQUIRESTARTED is sent during state 5 (source is ready to acquire), and capability negotiation is limited in states 5, 6, and 7 (only the caps that are extended caps are supported for states 5/6/7 negotiation).

[dmccallie]

We'd like to be able to set the Printer String for each image dynamically, per image. We'd know the right string right before the scan begins. Is that likely possible? When using callbacks, when is the device in state 4?

[dynarithmic]

We'd know the right string right before the scan begins. Is that likely possible?

Unless the CAP_PRINTER... capabilities are also extended capabilities (if the Source supports extended capabilities), negotiating these caps during states 5, 6, and 7 is not possible.

Going by the JSON listing, the CAP_PRINTER... capabilities are listed as only "standard", so they shouldn't be able to be negotiated while in states 5 - 7. So you have to negotiate these caps when the UI is closed, or if no UI, before the scan has started and after all pages have been scanned.

So in general try setting those caps,

1. before the DTWAIN_Acquire... call or after the call has returned, or
2. one of the callbacks listed in state 4 (below).

When using callbacks, when is the device in state 4?

DTWAIN_TN_ACQUIREDONE, DTWAIN_TN_ACQUIRECANCELED, DTWAIN_TN_UIOPENING and DTWAIN_TN_UICLOSED are the ones that should be sent when the source is in state 4.