Superfluous or Duplicate Files in Repository

[Bike]

Several files look like they're only included in the distribution accidentally.

• src/t.txt is some kind of clang session.
• src/chems is a tree structure of chemical names
• src/llvmo/ll.sh is a directory listing as from ls -l
• src/mpip/t.sh looks to be a command line from drmeister's personal environment
• src/asttooling/session.txt
• src/main/allHeaders_dump.cc* and unityBuild.cc* are series of includes of drmeister's absolute paths
• src/main/test.zzz
• src/main/t.sh
• src/core/exceptions.d has many drmeister absolute paths
• src/cando may be nanotech detritus
• src/main/all_debug.txt

will update if i notice more.

[fmitha]

src/core/floor2.txt <- looks like a C file? src/cando/a <- build attempt? Not sure what src/cando is for. Looks like a prototype attempt. Should probably be removed.

[erm410]

Also several symbols_scraped_inc.h (notably excluding the core one). They should probably all be included or all excluded. Since they are generated file my vote is for excluded. This also relates to issue #17

[drmeister]

Thank you very much. I removed all of these files from the preview/devel branches. I'll merge them tomorrow if an overnight build is successful.

[Bike]

trawling src/core

• .gdb reads "hello"
• _Function empty
• _bjam.txt build record
• _compile_te_cmp.txt build record
• _dl.cxml couldn't even guess
• _profile.txt presumably profiling data, which I don't think belongs here
• a.csc
• a.make build record
• a.sed
• aaa.dat empty
• all_release.txt empty
• am4 some kind of wxWidgets thing for cando?
• argv.csc
• array0.lsp test code
• atomicWeights.txt
• b.csc
• bottom.txt
• bt.txt
• build-stacks.sh compiler invocation, mentions cando
• buildMany.py mentions cyclomers
• c
• candoClass.cc unless integration with nanotech is deeper than i thought; it references apparently nonexistent file candoClass.h
• ceil2.txt, presumably ceiling is actually implemented elsewhere
• change.sed here's hoping sed isn't part of the build process
• changeAll ditto and also mentions oligomers
• chemInfoParse.output
• chems buncha filenames
• cl.sh compiler invocation mentions cando
• class.cc.prototype mentions cando
• class.h.prototype ditto
• classInitialization.txt is notes, might be worth keeping, i don't know, but probably not here
• compile is a build record
• compile-symbolList.sh build command mentions cando
• compare.txt some kind of test?
• convert-vp0.el
• copyright maybe fine?
• copyrightAll maybe fine?
• cores buncha filenames
• def.csc m expressions!
• dielsAlderCyclomersFinal.cgd always good to get a refresher on orgo
• dmake got your email address
• docCommands.aux
• docCommands.tex
• docFilter.m4
• docs.template scores?
• dox.out build record
• doxygen.out build record
• dumpClasses.lsp cannot imagine what this is for
• e.lsp part of a formatter?
• ecl.lisp test code
• editDocs.* whole buncha shit that might have been cando docs at some point?
• environment.h.old
• evaluateEnergy.cc I'll be excited if optimization involves hessians
• example.doc some interesting salespersonship for cando
• exportedSymbols.txt
• f.csc
• fact.l test code
• fixHeader again, really hoping this isn't part of build
• fixInitialize
• fixVariadicMacros.m4
• fixes
• fixes.txt
• forceField.list
• forceField.txt
• foundation.cc not sure, but references cando and boost::python?
• foundation.h.pth empty
• foundation.hh might be legit but mentions some chemistry stuff
• g.csc
• gaffParse.output
• gamess.cc
• gl.csc
• globalOptions.csc seriously, are these corelscript or what?
• grep
• headerFiles.txt
• i.txt
• implementAmberFunction.cc
• improperRestraintTerms.cc
• indent.csc
• initClasses.inc no idea but looks generated
• initializeClasses.cc misspells "deprecated"
• initializeClassesWithWx.cc
• interactions.txt
• inv.csc
• key.csc
• lambdaListHandler consists of "c", probably a typo
• mainBoostPython.cc dunno what the hell this is
• make.log.part is over a megabyte and probably a build record
• makeDatabase.py cyclomers
• makeInitPythonCalls.awk
• makeInitPythonDeclares.awk
• makeSed.awk frightening filename
• makefile* possibly cando stuff
• matter.cc.old
• mbb* chemistry stuff
• ml.sh build command
• modified_lispLex.yy
• mpi*
• ms* is stuff for parsing SMILES or an extension of it or something

need food before looking at the rest i'd like clasp to be understandable, and all these files floating around do not help that

[Shinmera]

There's probably new ones by now, so we should go through this again once there's a reproducible build.

[jorams]

Just to note some more things that seem useless, overlapping with @Bike's list:

• The src/tests/old directory
• Various .tex files and tex outputs in the src/core directory
• Various .txt files in the src/core directory
• A whole bunch of .l files in the src/core directory, that seem to have been used for testing
• A couple of [a-z]{1,3}.csc files in the src/core directory that are either useless or need a better name/location
• Various .old files, the old directory and the oldArchive directory in src/core
• All kinds of test*.* files in src/core
• src/core/{.gdb,timing,am4,a.make,a.sed,TODO,_Function,aaa.dat

[Shinmera]

Merging in from #45:

See the list of file listed in https://bitbucket.org/faheem/clasp-debian/src/tip/rules under the target override_dh_auto_clean: If these were moved into make clean, they could be removed from debian/rules.

[Shinmera]

Is this resolved now from the changes in dev-delete?

[Georgiy-Tugai]

I believe that this has been resolved; I will, however, quote rptb1's advice on the matter -- <rptb11> Make drmeister log in and get work done on a fresh account on a fresh computer :)