Harvesting requirements from uploaded documents?
Good day,
I have 3 questions, but I will create 3 different posts for each topic.
For this 1st question, we have a set of users that want to be able to upload documents into DOORS NG so that they can search requirements and harvest additional requirements.
I am thinking this sounds like something that would want to do in a tool other than DOORS NG. What is the best suggested solution for this scenario?
Thanks!
Mary
2 answers
Hi Mary,
There is a way that you can extract requirements from an uploaded text document after the fact, but in our case we either import to create the requirements at the start, or we set up a Supporting documents folder in which they can store uploaded docs. We find it works better to keep all related documents in the DNG project rather than in a separate repository, such as SharePoint. In some cases, an artifact might embed a link to one of the supporting documents.
Regards,
Carol
Comments
Okay, that is interesting. I wasn't sure what standard practices are used.
For the way of extracting requirements from an uploaded text document after the fact, would this be a DOORS NG functionality or would I have to script it?
Do you modify the document ahead of time to have shall statements and then import it?
Thanks again!
Mary
Extracting is a feature, so no script needed. If you're using Modules (requirements where hierarchy is desired), DNG uses the Word Headings to determine what to make a 'numbered' heading. From there it pretty much makes every sentence a separate artifact, unless you used bullets or numbering, in which case it will put that "chunk" in one artifact.
You can import from Excel too. That works a bit better if you're using Attributes, since you can import the columns into the Attributes. It too can be used to create a Module, it's just a bit more work identifying the hierarchy prior to importing (it's uses something called ParentBinding).
In our case, if they've been using Word or Excel already, we try to make the most of what they already have documented.
I'd suggest getting a trial sandbox to play around, that's what we did and it was great.
Carol
Okay, I figured that is what was needed, and I wanted to make sure I wasn't missing anything. Thank you for the clarification!
Mary
If you are looking for automated extraction (particularly if the source documents are in PDF format), you could look at https://www.em.ag/reqman/?lang=en
ReqMan provides a way to transform docs into machine-ready requirement. After transformation you export a ReqIF file that can be imported into DNG as native requirements. Their web site frames the value in terms of accelerating responses to RFPs, which is a valid use case but by no means the only one.