--
You received this message because you are subscribed to a topic in the Google Groups "JCrypTool Developers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/jcryptool-developers/wf7XwELDX-g/unsubscribe.
To unsubscribe from this group and all its topics, send an email to jcryptool-develo...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
To unsubscribe from this group and all its topics, send an email to jcryptool-developers+unsub...@googlegroups.com.
Hi Bruno,thanks for the compliment regarding the substitution analysis :)Well, as you noted, there are some things that could be improved. Indeed, predefined statistics in the context do not mean that the statistical distributions themselves have been calculated already. They are calculated each time an analysis is queried. Perhaps, to clarify what is being done, let me point out a few things:
- the distribution of 1- 2- and triplets depend on the alphabet selected, e.g. if you choose to only analyze for lower-case characters by choosing an according alphabet, the doublet "Cr" from "JCrypTool" would not appear. It would be possible to precalculate and store statistics for a reference text for the biggest-possible alphabet there is and marginalize the distribution based on the alphabet later, but that was too convuluted a concept at that time. Note also, that the largest-possible alphabet would include every character from the UTF-8 character set, which is quite large.- German for the one purpose is not German for another purpose, as is with any other language. "The murders in the rue morgue" is a short story written in perhaps dated English. The statistics for a football match transcription may look entirely different, containing much more digits, as would cooking recipes etc. Therefore, the concept of "language to analyze" was conveniently tacked onto a reference text which represents the "true" reference in the context of the analysis. Effectively, the actually selected reference text IS the language that is analyzed.This actually makes it easier not harder, to implement a french analysis!All that is needed to make the substitution analysis "understand french" is to1) put a french reference text in the plugin's folder "refTexts" - https://github.com/jcryptool/crypto/tree/develop/org.jcryptool.analysis.substitution/refTexts2) add it to the list of reference texts the plugin knows here: https://github.com/jcryptool/crypto/blob/develop/org.jcryptool.analysis.substitution/src/org/jcryptool/analysis/substitution/calc/DynamicPredefinedStatisticsProvider.java#L78-L86For minimum success, all that is needed to put these two lines there:
String dumas = "refTexts/Le_Comte_de_Monte-Cristo.txt"; //$NON-NLS-1$predefinedTextFiles.put(new Info("Le Comte de Monte-Cristo (Dumas)", "French"), dumas);and it would already work.Then, you could translate the String "french" as seen with the other reference texts, by invoking the eclipse tool for internationalization (Right cick in the source code, -> "Source" -> "Externalize Strings"), then check the fileshttps://github.com/jcryptool/crypto/blob/develop/org.jcryptool.analysis.substitution/src/org/jcryptool/analysis/substitution/calc/messages.properties and
https://github.com/jcryptool/crypto/blob/develop/org.jcryptool.analysis.substitution/src/org/jcryptool/analysis/substitution/calc/messages_de.propertiesfor completeness of that translation. It's not perfectly necessary to do that right away yourself, we don't shun contributions if they are not perfectly translated ;)I hope that clears up the important parts.
To unsubscribe from this group and all its topics, send an email to jcryptool-developers+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to a topic in the Google Groups "JCrypTool Developers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/jcryptool-developers/wf7XwELDX-g/unsubscribe.
To unsubscribe from this group and all its topics, send an email to jcryptool-developers+unsub...@googlegroups.com.
To unsubscribe from this group and all its topics, send an email to jcryptool-develo...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to a topic in the Google Groups "JCrypTool Developers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/jcryptool-developers/wf7XwELDX-g/unsubscribe.
To unsubscribe from this group and all its topics, send an email to jcryptool-develo...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to a topic in the Google Groups "JCrypTool Developers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/jcryptool-developers/wf7XwELDX-g/unsubscribe.
To unsubscribe from this group and all its topics, send an email to jcryptool-develo...@googlegroups.com.
To unsubscribe from this group and all its topics, send an email to jcryptool-developers+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to a topic in the Google Groups "JCrypTool Developers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/jcryptool-developers/wf7XwELDX-g/unsubscribe.
To unsubscribe from this group and all its topics, send an email to jcryptool-developers+unsub...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to a topic in the Google Groups "JCrypTool Developers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/jcryptool-developers/wf7XwELDX-g/unsubscribe.
To unsubscribe from this group and all its topics, send an email to jcryptool-developers+unsub...@googlegroups.com.