Search found 1225 matches

by Marco Eichelberg
Fri, 2018-11-30, 15:43
Forum: DCMTK - General
Topic: ftoa incorrect conversion of numeric_limits<double>::max
Replies: 1
Views: 61

Re: ftoa incorrect conversion of numeric_limits<double>::max

Thanks for the report. I have logged this in the DCMTK issue tracker as issue 860.
by Marco Eichelberg
Sun, 2018-11-25, 11:54
Forum: DCMTK - General
Topic: Accepted transfer syntax
Replies: 1
Views: 783

Re: Accepted transfer syntax

I guess the answer comes a bit late, but it's actually quite simple. The algorithm will accept the first transfer syntax from your own list of preferences that was proposed by the caller, so in this case, the result would be LittleEndianExplicit.
by Marco Eichelberg
Sun, 2018-11-25, 11:46
Forum: DCMTK - General
Topic: Questions about timeouts.
Replies: 2
Views: 137

Re: Questions about timeouts.

The ARTIM timer is described in detail in DICOM part 9 and only handles very specific timeout conditions: A network connection was received but no A-ASSOCIATE-RQ was received before the ARTIM timer expires An association was rejected (A-ASSOCIATE-RQ) and the remote side has not closed the connection...
by Marco Eichelberg
Sun, 2018-11-25, 11:37
Forum: DCMTK - General
Topic: Bug #373
Replies: 1
Views: 130

Re: Bug #373

Thanks for the information. Your description of the condition under which the bug occurs should be precise enough to allow us to fix this. This will not make it into DCMTK 3.6.4, though, which is due for release in a couple of days.
by Marco Eichelberg
Sun, 2018-11-25, 11:33
Forum: DCMTK - General
Topic: TLS standard version when OpenSSL version < 1.1.0
Replies: 2
Views: 129

Re: TLS standard version when OpenSSL version < 1.1.0

Brief Answer: In the public git repository of the DCMTK project you can find a completely revised implementation of the TLS support that will be released in the upcoming DCMTK 3.6.4 release. Check how we have approached this topic there. Long Answer: You cannot use the TLS_* methods with OpenSSL 1.0...
by Marco Eichelberg
Thu, 2018-11-01, 12:48
Forum: DCMTK - Installation
Topic: Problem building with OpenSSl support on macOS
Replies: 8
Views: 259

Re: Problem building with OpenSSl support on macOS

It is true that DCMTK up to release 3.6.2 compiled with very old OpenSSL releases. The TLS code in DCMTK has been completely revised recently, implementing newer security profiles meanwhile defined in the DICOM standard. In particular, we now require TLS 1.2 which in turn needs OpenSSL 1.0.2. DCMTK ...
by Marco Eichelberg
Thu, 2018-11-01, 12:38
Forum: DCMTK - General
Topic: movescu can not retrieve image file
Replies: 15
Views: 2152

Re: movescu can not retrieve image file

Also note: the retrieve request is not well-formed. A study level retrieve request in the Study Root Q/R model must not contain a Patient ID.
by Marco Eichelberg
Tue, 2018-10-23, 12:28
Forum: DCMCHECK
Topic: Updating the Data Dictionary of dcmcheck
Replies: 2
Views: 136

Re: Updating the Data Dictionary of dcmcheck

Exactly. DCMCHECK 2.0 is based on DCMTK 3.5.2 and does not yet know about VRs like UC, UR or OL. If you remove these tags from the data dictionary, this will be one step forward. There is one more thing you will have to do: DCMCHECK checks if the readable attribute names in its script files are cons...
by Marco Eichelberg
Tue, 2018-09-04, 11:30
Forum: DCMTK - General
Topic: Signed Short Pixel Representation sample file
Replies: 1
Views: 284

Re: Signed Short Pixel Representation sample file

1. Changing just the pixel representation = 1 should NOT ideally work as the pixel itself has to be now in Signed Short format right ? In a Basic Grayscale Image Box N-SET-RQ, Pixel Representation is required to have the value 0 (as per the definition of the service in DICOM part 4). DCMTK checks t...
by Marco Eichelberg
Mon, 2018-09-03, 19:26
Forum: DCMPRINT
Topic: Re-printing stored print objects
Replies: 13
Views: 1435

Re: Re-printing stored print objects

Both the DICOM print implementation in the free toolkit, and DCMPRINT have the ability to store incoming print jobs using the DICOM Stored Print and Hardcopy Grayscale/Color Image SOP Classes. When you are using the DCMPRINT print server (tcpsrv), you can use the --stored-print and --forward command...
by Marco Eichelberg
Sun, 2018-07-29, 09:39
Forum: DCMTK - General
Topic: Cannot determine start fragment: basic offset table is empty
Replies: 4
Views: 518

Re: Cannot determine start fragment: basic offset table is e

I also had a look at the image. The reason why direct frame-wise access did not work is actually that the image is faulty: The image contains 139 frames, but NumberOfFrames is only 138. As you perhaps know, the DICOM standard allows JPEG compressed data to be store in a fragmented manner, i.e. there...
by Marco Eichelberg
Tue, 2018-04-03, 08:37
Forum: DCMTK - General
Topic: bug in poll/connect
Replies: 2
Views: 524

Re: bug in poll/connect

Thanks for your posting. We are aware of the pull request, and it's on the "to do list" (I think Michael Onken is working on it currently).
by Marco Eichelberg
Mon, 2018-02-19, 15:52
Forum: DCMTK - General
Topic: dcmj2pnm: RLE Lossless problem with US Singleframe Image
Replies: 6
Views: 958

Re: dcmj2pnm: RLE Lossless problem with US Singleframe Image

No, I don't think so. The error is somewhere in the middle of the RLE bytestream, and I cannot imagine any network receiver or CD/DVD burning tool to mess with this at this low level. My best guess is that the invalid RLE bytestream comes directly from the modality. The easiest way of avoiding that ...
by Marco Eichelberg
Mon, 2018-02-19, 15:48
Forum: DCMTK - General
Topic: Open SSL Cipher Suites
Replies: 10
Views: 1640

Re: Open SSL Cipher Suites

Well, at this point you'll probably have to check things with the debugger. I noticed that the new cipher suites you refer to are TLSv1.2 suites. You should probably check which TLS version is proposed by the client, and which version(s) the server is willing to accept. The error code "unspecified T...
by Marco Eichelberg
Sun, 2018-02-18, 12:19
Forum: DCMTK - General
Topic: dcmj2pnm: RLE Lossless problem with US Singleframe Image
Replies: 6
Views: 958

Re: dcmj2pnm: RLE Lossless problem with US Singleframe Image

Thanks for the sample image. The patch you refer is indeed in the code, but does not help here. The workaround to fill in the last missing pixels with the value of the last pixel only works at the very end of the pixel data element, i.e. if the pixel data element is for some reason truncated. In thi...