Re: [Jmol-users] Important Java changes - January 2014
Hi Bob, I know I have a full certificate, valid for one year, but it's unclear what I will have to do in one year : I'm not sure if renewal is still free or requires some money. It's the only option I found that allowed me to have a free code signing certificate for the moment, so I decided to try it. Users with recent Java version have now again the ability to trust my application once and for all, which is a lot better than what they were allowed with the self-signed certificate. Nico On Sun, Nov 3, 2013 at 12:51 AM, Robert Hanson hans...@stolaf.edu wrote: hang in there. I think we will be ok. But, Nico, are you sure you don't have a 'trial version' On Sat, Nov 2, 2013 at 2:20 PM, Nicolas Vervelle nverve...@gmail.comwrote: Hi Jonathan, I think we can try using free certificates, I've got one 2 weeks ago from certum as they give them for free to open source developers. I'm using it for a Java application, and it seems to work a lot better than with the self signed one. If need be, I can sign a Jmol version with it to see if it works correctly with Jmol Nico On Sat, Nov 2, 2013 at 7:49 PM, Jonathan Gutow gu...@uwosh.edu wrote: Ouch! I've been out of the loop for a while trying to get caught up on local work issues. This does not sound good. As I read the description from Oracle the only way to use Java in browsers that does not require a commercial certificate and a codebase compiled for each server (possibly you don't have to specify your codebase location in the jar, but I think you do) is to have each local user/intranet whitelist the particular jar from a particular source. This may be OK within a large company infrastructure, but is not going to work for general users. Have I misread this? The most amusing thing about this is that the alternative (javascript and things like opengl) are no more secure for the same capabilities and so far much slower. We still need the java application, but I guess we pretty much need to get everything converted to JSmol for web stuff. Jonathan On Nov 2, 2013, at 1:11 PM, jmol-users-request@lists.sourceforge.netwrote: Assuming the $500 certificate ensures that the signed Jmol java applet will not be blocked (?), I suspect there are a number of organizations that would be prepared to become sponsors... Quoting Robert Hanson hans...@stolaf.edu: I direct the discussion to https://blogs.oracle.com/java-platform-group/entry/new_security_requirements_for_rias I believe this is the end of the unsigned Jmol Java applet along with JSpecView and JME. In addition, I'm pretty sure our free-be signing will not pass muster as a trusted authority: RIAs must contain two things: 1. Code signatures from a trusted authority. All code for Applets and Web Start applications must be signed, regardless of its Permissions attributes. 2. Manifest Attributes 1. Permissions ? Introduced in 7u25, and required as of 7u51. Indicates if the RIA should run within the sandbox or require full-permissions. 2. Codebase ? Introduced in 7u25 and optional/encouraged as of 7u51. Points to the known location of the hosted code (e.g. intranet.example.com) http://intranet.example.com. The latest upload of Jmol takes care of (2a). However, unless (2b) allows Codebase: * that's pretty much it for the signed applet as well. [Or maybe someone goes into the business of making custom signed Jmol applets for people!] Suggestions? Comments? If deployment of the signed Jmol applet is of interest, we will need a sponsor, because a certificate costs US$500/year. Let me know if you are interested in being that sponsor. At least we have a two-month lead on this (and I am headed for a visit with RCSB on Sunday). Bob Dr. Jonathan H. Gutow Chemistry Departmentgu...@uwosh.edu UW-Oshkosh Office: 920-424-1326 800 Algoma BoulevardFAX:920-424-2042 Oshkosh, WI 54901 http://www.uwosh.edu/facstaff/gutow -- Android is increasing in popularity, but the open development platform that developers love is also attractive to malware creators. Download this white paper to learn more about secure code signing practices that can help keep Android apps secure. http://pubads.g.doubleclick.net/gampad/clk?id=65839951iu=/4140/ostg.clktrk ___ Jmol-users mailing list Jmol-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jmol-users -- Android is increasing in popularity, but the open development platform that developers love is also attractive to malware creators. Download this white paper to learn
Re: [Jmol-users] Important Java changes - January 2014
Excellent. Will this also be true for the applet version? In one year from now we’ll see if Java's still required, and, if the renewal requires a payment, find sources for that. Jaim __ Dr Jaime Prilusky Head Bioinformatics RD Bioinformatics and Data Management Department of Biological Services Weizmann Institute of Science 76100 Rehovot - Israel mail: jaime.prilu...@weizmann.ac.ilmailto:jaime.prilu...@weizmann.ac.il tel: 972-8-9344959 fax: 972-8-9344113 OCA, http://oca.weizmann.ac.il (the protein structure/function database) Proteopedia, http://proteopedia.org (because life has more than 2D) On Nov 3, 2013, at 10:57 AM, Nicolas Vervelle nverve...@gmail.commailto:nverve...@gmail.com wrote: Hi Bob, I know I have a full certificate, valid for one year, but it's unclear what I will have to do in one year : I'm not sure if renewal is still free or requires some money. It's the only option I found that allowed me to have a free code signing certificate for the moment, so I decided to try it. Users with recent Java version have now again the ability to trust my application once and for all, which is a lot better than what they were allowed with the self-signed certificate. Nico On Sun, Nov 3, 2013 at 12:51 AM, Robert Hanson hans...@stolaf.edumailto:hans...@stolaf.edu wrote: hang in there. I think we will be ok. But, Nico, are you sure you don't have a 'trial version' On Sat, Nov 2, 2013 at 2:20 PM, Nicolas Vervelle nverve...@gmail.commailto:nverve...@gmail.com wrote: Hi Jonathan, I think we can try using free certificates, I've got one 2 weeks ago from certum as they give them for free to open source developers. I'm using it for a Java application, and it seems to work a lot better than with the self signed one. If need be, I can sign a Jmol version with it to see if it works correctly with Jmol Nico On Sat, Nov 2, 2013 at 7:49 PM, Jonathan Gutow gu...@uwosh.edumailto:gu...@uwosh.edu wrote: Ouch! I've been out of the loop for a while trying to get caught up on local work issues. This does not sound good. As I read the description from Oracle the only way to use Java in browsers that does not require a commercial certificate and a codebase compiled for each server (possibly you don't have to specify your codebase location in the jar, but I think you do) is to have each local user/intranet whitelist the particular jar from a particular source. This may be OK within a large company infrastructure, but is not going to work for general users. Have I misread this? The most amusing thing about this is that the alternative (javascript and things like opengl) are no more secure for the same capabilities and so far much slower. We still need the java application, but I guess we pretty much need to get everything converted to JSmol for web stuff. Jonathan On Nov 2, 2013, at 1:11 PM, jmol-users-requ...@lists.sourceforge.netmailto:jmol-users-requ...@lists.sourceforge.net wrote: Assuming the $500 certificate ensures that the signed Jmol java applet will not be blocked (?), I suspect there are a number of organizations that would be prepared to become sponsors... Quoting Robert Hanson hans...@stolaf.edumailto:hans...@stolaf.edu: I direct the discussion to https://blogs.oracle.com/java-platform-group/entry/new_security_requirements_for_rias I believe this is the end of the unsigned Jmol Java applet along with JSpecView and JME. In addition, I'm pretty sure our free-be signing will not pass muster as a trusted authority: RIAs must contain two things: 1. Code signatures from a trusted authority. All code for Applets and Web Start applications must be signed, regardless of its Permissions attributes. 2. Manifest Attributes 1. Permissions ? Introduced in 7u25, and required as of 7u51. Indicates if the RIA should run within the sandbox or require full-permissions. 2. Codebase ? Introduced in 7u25 and optional/encouraged as of 7u51. Points to the known location of the hosted code (e.g. intranet.example.comhttp://intranet.example.com/) http://intranet.example.comhttp://intranet.example.com/. The latest upload of Jmol takes care of (2a). However, unless (2b) allows Codebase: * that's pretty much it for the signed applet as well. [Or maybe someone goes into the business of making custom signed Jmol applets for people!] Suggestions? Comments? If deployment of the signed Jmol applet is of interest, we will need a sponsor, because a certificate costs US$500/year. Let me know if you are interested in being that sponsor. At least we have a two-month lead on this (and I am headed for a visit with RCSB on Sunday). Bob Dr. Jonathan H. Gutow Chemistry Department gu...@uwosh.edumailto:gu...@uwosh.edu UW-Oshkosh Office: 920-424-1326tel:920-424-1326 800 Algoma Boulevard
Re: [Jmol-users] Important Java changes - January 2014
Hi, Yes, I think it's the same for applets as for Java Web Start applications. I haven't tested, but I see no reasons for it to be different. Nico On Sun, Nov 3, 2013 at 10:17 AM, Jaime Prilusky jaime.prilu...@weizmann.ac.il wrote: Excellent. Will this also be true for the applet version? In one year from now we’ll see if Java's still required, and, if the renewal requires a payment, find sources for that. Jaim __ Dr Jaime Prilusky Head Bioinformatics RD Bioinformatics and Data Management Department of Biological Services Weizmann Institute of Science 76100 Rehovot - Israel mail: jaime.prilu...@weizmann.ac.il tel: 972-8-9344959 fax: 972-8-9344113 OCA, http://oca.weizmann.ac.il (the protein structure/function database) Proteopedia, http://proteopedia.org (because life has more than 2D) On Nov 3, 2013, at 10:57 AM, Nicolas Vervelle nverve...@gmail.com wrote: Hi Bob, I know I have a full certificate, valid for one year, but it's unclear what I will have to do in one year : I'm not sure if renewal is still free or requires some money. It's the only option I found that allowed me to have a free code signing certificate for the moment, so I decided to try it. Users with recent Java version have now again the ability to trust my application once and for all, which is a lot better than what they were allowed with the self-signed certificate. Nico On Sun, Nov 3, 2013 at 12:51 AM, Robert Hanson hans...@stolaf.edu wrote: hang in there. I think we will be ok. But, Nico, are you sure you don't have a 'trial version' On Sat, Nov 2, 2013 at 2:20 PM, Nicolas Vervelle nverve...@gmail.comwrote: Hi Jonathan, I think we can try using free certificates, I've got one 2 weeks ago from certum as they give them for free to open source developers. I'm using it for a Java application, and it seems to work a lot better than with the self signed one. If need be, I can sign a Jmol version with it to see if it works correctly with Jmol Nico On Sat, Nov 2, 2013 at 7:49 PM, Jonathan Gutow gu...@uwosh.edu wrote: Ouch! I've been out of the loop for a while trying to get caught up on local work issues. This does not sound good. As I read the description from Oracle the only way to use Java in browsers that does not require a commercial certificate and a codebase compiled for each server (possibly you don't have to specify your codebase location in the jar, but I think you do) is to have each local user/intranet whitelist the particular jar from a particular source. This may be OK within a large company infrastructure, but is not going to work for general users. Have I misread this? The most amusing thing about this is that the alternative (javascript and things like opengl) are no more secure for the same capabilities and so far much slower. We still need the java application, but I guess we pretty much need to get everything converted to JSmol for web stuff. Jonathan On Nov 2, 2013, at 1:11 PM, jmol-users-request@lists.sourceforge.netwrote: Assuming the $500 certificate ensures that the signed Jmol java applet will not be blocked (?), I suspect there are a number of organizations that would be prepared to become sponsors... Quoting Robert Hanson hans...@stolaf.edu: I direct the discussion to https://blogs.oracle.com/java-platform-group/entry/new_security_requirements_for_rias I believe this is the end of the unsigned Jmol Java applet along with JSpecView and JME. In addition, I'm pretty sure our free-be signing will not pass muster as a trusted authority: RIAs must contain two things: 1. Code signatures from a trusted authority. All code for Applets and Web Start applications must be signed, regardless of its Permissions attributes. 2. Manifest Attributes 1. Permissions ? Introduced in 7u25, and required as of 7u51. Indicates if the RIA should run within the sandbox or require full-permissions. 2. Codebase ? Introduced in 7u25 and optional/encouraged as of 7u51. Points to the known location of the hosted code (e.g. intranet.example.com) http://intranet.example.com. The latest upload of Jmol takes care of (2a). However, unless (2b) allows Codebase: * that's pretty much it for the signed applet as well. [Or maybe someone goes into the business of making custom signed Jmol applets for people!] Suggestions? Comments? If deployment of the signed Jmol applet is of interest, we will need a sponsor, because a certificate costs US$500/year. Let me know if you are interested in being that sponsor. At least we have a two-month lead on this (and I am headed for a visit with RCSB on Sunday). Bob Dr. Jonathan H. Gutow Chemistry Departmentgu...@uwosh.edu UW-Oshkosh
Re: [Jmol-users] I got it
Bob,
I took the if statement out and ran this:
set Picking OFF;set Picking ON;set BondPicking TRUE;set allowRotateSelected
TRUE;unbind SHIFT-SINGLE-LEFT-DRAG;bind SHIFT-SINGLE-LEFT-DRAG a1 =
{atomIndex=rot1};a2 = {atomIndex=rot2};b1 = {WITHIN(BRANCH,a1,a2)};b2 =
{WITHIN(BRANCH,a2, a1)}; select b1;rotateSelected @a1 @a2 @{_DELTAY};hover off;
It still freezes randomly with console simply terminating at the final pre
freeze:
a1 = {atomIndex=rot1};
a2 = {atomIndex=rot2};
b1 = {WITHIN(BRANCH,a1,a2)};
b2 = {WITHIN(BRANCH,a2, a1)};
select b1;
rotateSelected @a1 @a2 @{-44};
I went back to 13.3.3 and checked the above with old bind syntax. I also had
another bind script that I tested on 13.3.3:
set picking off; set picking ON; set atomPicking true; set allowRotateSelected
TRUE; set picking dragmolecule; unbind DOUBLE-DRAG; bind DOUBLE-DRAG
rotateSelected Y @{_DELTAX*1.5};rotateSelected X @{_DELTAY*1.5}
Both freeze randomly. These used to work on Safari - even on iPad. So going
back to 13.3.3 tells me this incompletely a Safari issue.
The recent upgrade to Apple Mavericks OS was not a pleasant experience. Safari
iPad crashes frequently.
I think I'm going to have to find a non callback way of dealing with this. One
thing that's a bit of a bummer is that other callbacks (model editing) are also
freezing, albeit less frequently. Again, these are all Safari Mavericks issues.
One option is to turn modelkitmode on and off as needed. A long time ago you
sent me an email with hooks for controlling basic edit operations for modelkit
mode in script. Any chance you still have that email?
Long Post! Bottom line: Mavericks OS JSmol scripters keep your eyes open for
problems.
Otis
--
Otis Rothenberger
o...@chemagic.com
http://chemagic.com
On Nov 3, 2013, at 12:21 AM, Robert Hanson hans...@stolaf.edu wrote:
another thought: This is the sort of error that would come from
multithreading -- two threads processing the same code block at the same
time. That's not supposed to be possible in JavaScript, but. maybe
Safari
--
Android is increasing in popularity, but the open development platform that
developers love is also attractive to malware creators. Download this white
paper to learn more about secure code signing practices that can help keep
Android apps secure.
http://pubads.g.doubleclick.net/gampad/clk?id=65839951iu=/4140/ostg.clktrk
___
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users
[Jmol-users] Modelkitmode
Bob, I'm thinking about redesigning my interface to use modelkitmode for the model editing stuff. I know I can invoke modelkitmode via script. Is there a way to also bring up the popup modelkitmode menu? Could this menu be made touch screen friendly? Otis -- Otis Rothenberger o...@chemagic.com http://chemagic.com -- Android is increasing in popularity, but the open development platform that developers love is also attractive to malware creators. Download this white paper to learn more about secure code signing practices that can help keep Android apps secure. http://pubads.g.doubleclick.net/gampad/clk?id=65839951iu=/4140/ostg.clktrk___ Jmol-users mailing list Jmol-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jmol-users
Re: [Jmol-users] R: Re: Parsing JME string
load SMILES xxx is what I am using. However what I found it, the smiles string is not sufficient to depict the original 2D structure. Some of the molecules we tried to build lost the double bonds when we get the result from the cactus. That's why I am search for alternative solutions. So I am thinking of using the 2D representation of the structures e.g. the sdf file, but I was not able to foresee the conversion of the sdf file to other formats could be so troublesome. I thought it is straightforward in babel. However now I am stuck at the adding correct number of H atoms to the charged atoms in sdf file which is kind of disturbing. Anyway, I still am looking for ways to correct the number of H atoms. cheers, _ Wenhan Chen School of Chemistry and Molecular Biosciences The University of Queensland 2013/11/1 Robert Hanson hans...@stolaf.edu The CACTVS server at NIH does a great job of turning 2D files into 3D. It's best not to have Jmol do that. You can pass the SMILES string from JME to Jmol; it will send it off to NIH. load SMILES xxx On Tue, Oct 29, 2013 at 7:55 PM, 陈文瀚 chenwh...@gmail.com wrote: Actually if I transfer the structure from a mol file to JSmol and then try to optimize the structure, the charity cannot be preserved as well ... _ Wenhan Chen School of Chemistry and Molecular Biosciences The University of Queensland 2013/10/30 陈文瀚 chenwh...@gmail.com Basically, molecules with more than 30 atoms sent in JME format(not including H) will cause a halt for the browser as the js try to hold the browser. But you are right, it was quick using Jmol with JRE. I think JSmol try to do EM using MMFF which takes the most time. When the molecule in MOL format passes onto JSmol, it works fine with no EM. Even so I still am trying to address the problem using mol format, as the charge information (the number of H atoms) in the mol file is not presented right in JSmol. Herein is an example : C+ / \ C -- -C The mol file for the above mol is : CWRITER31030r]09392D Created with ChemWriter - http://chemwriter.com 3 3 0 0 0 0 0 0 0 0999 V2000 4.3530 -4.78300. C 0 0 0 0 0 0 0 0 0 0 0 0 5.3530 -4.78300. C 0 0 0 0 0 0 0 0 0 0 0 0 4.8530 -3.91690. C 0 0 0 0 0 0 0 0 0 0 0 0 1 2 1 0 0 0 2 3 1 0 0 0 3 1 1 0 0 0 M CHG 1 3 1 M END JSmol cannot add an H atom to the C+ atom. _ Wenhan Chen School of Chemistry and Molecular Biosciences The University of Queensland 2013/10/30 Angel Herráez angel.herr...@uah.es Thanks for the feedback, Pino Then maybe the 3D minimization is much slower in the Javascript version that in the Java version. I've now added an option to the JSME page to check if it is the transfer (unlikely), adding the hydrogens or the 3D minimization which gives the delay That is only useful for MOL transfer. JME transfer must necessarily go to 3D conversion, that may be the bottleneck. It seems indeed that it is the JmeReader routine that takes its time -- Android is increasing in popularity, but the open development platform that developers love is also attractive to malware creators. Download this white paper to learn more about secure code signing practices that can help keep Android apps secure. http://pubads.g.doubleclick.net/gampad/clk?id=65839951iu=/4140/ostg.clktrk ___ Jmol-users mailing list Jmol-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jmol-users -- Android is increasing in popularity, but the open development platform that developers love is also attractive to malware creators. Download this white paper to learn more about secure code signing practices that can help keep Android apps secure. http://pubads.g.doubleclick.net/gampad/clk?id=65839951iu=/4140/ostg.clktrk ___ Jmol-users mailing list Jmol-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/jmol-users -- Robert M. Hanson Larson-Anderson Professor of Chemistry St. Olaf College Northfield, MN http://www.stolaf.edu/people/hansonr If nature does not answer first what we want, it is better to take what answer we get. -- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900 -- Android is increasing in popularity, but the open development platform that developers love is also attractive to malware creators. Download this white paper to learn more about secure code signing practices that
