-
Notifications
You must be signed in to change notification settings - Fork 37
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Symmetry operation specifications in Jones faithful notation (defined by PCRE) #464
Conversation
tools can be compiled and grammars can be tested on newer Java distributions.
match the documented REs.
NB: the text will be added/expanded later, for now its just the REs and the tests. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The regular expression for symmetry operators looks good to me. Text tying them to OPTIMADE properties is needed, but the regex part seems appropriate.
Workshop suggestion: we accept the syntax specification. The remaining fields from #416 #416 (comment) should be added to the PR and time given to symmetry description stakeholders to review. @rartino @sauliusg @vaitkus @merkys @giovannipizzi (feel free to post a comment below to add yourself to the stakeholders) |
trying to guess what the heck was meant by them... Removing the only trailing space in '.words.lst', which, bu the way, is standardly inserted by 'aspell' on linux :/.
To make it clear – after the rephrasing the text no longer refers to any property, either existing or not. The "fractional atom coordinates" expression that is used in the definition refers to a general crystallographic values. If there is a property in OPTIMADE that returns these values, the property can be used. If there is no such property, the fractional coordinates need to be computer in one way or another, and it is up to the client to do so. I've made this change in the text on purpose, so that it remains correct regardless whether OPTIMADE defines a fractional coordinate property or not. On this way we can merge this PR now and will not have to return to it when the fractional coordinate property is actually defined. To clarify the definition, I have added an example on how to deal with the situation when only Cartesian coordinates are returned in a response (see the sentence cited below).
In this case the reader is referred to standard textbooks on crystallography. Should we give a reference?
I cannot find the place where I mention application of symmetry operations to Cartesian coordinates; that would be also incorrect. Could you please give me the exact location of the text that you cite?
I can not find any clash here; as said, I can not find the place where Cartesian coordinates are mentioned as suitable for symop applications. Thus there should be no clash or contradiction here. I have added an explanation of how Cartesian coords could be used with symmetry operations: "In case only Cartesian coordinates are available, these Cartesian coordinates must be converted to fractional coordinates before application of the provided symmetry operations." . How about that?
I disagree. The symops as given can be applied to fractional coordinates. They must be applied if fractional coordinates are given for the AU only and the user wants to reconstruct the full unit cell. Both cases are relevant. |
@sauliusg wrote:
I was referring to part of the comment from (#464 (comment)):
But this has now been addressed with an additional sentence in the PR.
Then it should be "which these operations can or must be applied" instead of "which these operations can and must be applied" ("and" -> "or"). |
OK, so this is hopefully resolved now.
Good suggestion, thanks, I will apply it! |
…be applied" action is mentioned in the revious sentence... accepted. Co-authored-by: Antanas Vaitkus <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Comments during meeting.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
One more edit from meeting.
Co-authored-by: Rickard Armiento <[email protected]>
Co-authored-by: Rickard Armiento <[email protected]>
Applying suggestions discussed in the meeting. Co-authored-by: Rickard Armiento <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good now!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks great to me, thanks for the hard work!
Let me also add a direct link to the CIF dictionary mentioned. Also I’d like to check, if the ops non-symbolic notation is discouraged (e.g. |
@blokhin wrote:
A link to the https://www.iucr.org/__data/iucr/cifdic_html/1/cif_core.dic/Ispace_group_symop_operation_xyz.html The non-symbolic notation is not allowed at all by the regular expression, that is, |
Please have a look at the symmetry operation specification PR with the Jones' faithful notation of the symmetry operations.
Word description was added on top of formal PCRE definitions of the syntax.
To specify the semantics (interpretation) of the symmetry operations, I had to cite very specific places of the ITC vols. A and B. As a result, there are many references in the text now.
For now, I put the references after each definition section; this however results in scattered bibliography list and duplications (one duplication is present even now). I think it would be better to make a separate "Bibliography" section at the end of the text and place all bibliography sources there, with just references inserted into the text, using the Harvard referencing style (Author-year). If you agree (or are not much against this suggestion), I can rearrange references in this or in a separate PR.
I had to update also the definition of the Hall symbol; to be machine-parsable it needs to be more specific and allow the change-of-basis operations (COD has such entries).
There is currently no discussion on the rationale and intended use of H-M and Hall symbols. Basically the short Hall symbols are in 1:1 correspondence with the ITC space group numbers and fail to provide the setting and the cell choice information. The benefit of the H-M symbols however is that they are much more human-readable and allow to understand the properties of the space group without even looking into the ITC (unlike the pure ITC numbers). Good for searches.
The full (extended) H-M symbols, as well as Hall symbols (both with the c-o-b operations specified) should permit unique identification of the symmetry operation list (Hall symbols after algorithmic interpretation, H-M symbols after table lookup). If you think that this explanation would be needed in the definition text, please let me know, I'll add it.