Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with
or
.
Download ZIP
Browse files

rm TODO, here we have tickets

  • Loading branch information...
commit 9a671ba2c82d3f7ba0d3e7e386258f5c911992ec 1 parent 8227f55
@wactbprot authored
View
BIN  R4CouchDB_0.062.tar.gz
Binary file not shown
View
BIN  R4CouchDB_0.063.tar.gz
Binary file not shown
View
4 README
@@ -3,7 +3,7 @@ Package: R4CouchDB
Type: Package
Title: Collection of R functions for CouchDB access
Version: 0.07
-Date: 2011/06/26
+Date: 2011/09/13
Author: wactbprot
Maintainer: <thsteinbock|at|web.de>
Description: The R4CouchDB package provides a collection of functions for
@@ -14,7 +14,7 @@ Description: The R4CouchDB package provides a collection of functions for
"localhost" as a default. This means that localhost should
provide a CouchDB. The example sections in the Rd files (man
section) run with
- http://wactbprot.couchone.com.
+ http://wactbprot.couchone.com/_utils
This allows to provide (more or less) real examples.
Many thanx to the couchone relaxers
View
139 TODO
@@ -1,139 +0,0 @@
-- writing Tests -> RUnit package provides infrastructure
--- this should be the next step
---- test structure implemented; some test available; results go to the log-folder
-
-- before returning the list the answer should be more evaluated
-
-- function for view requests
--- done, but without params as ..?group=true
---- done
-
-- sending _design/_/_view/_ by means of Duncans RJavascript
- http://www.omegahat.org/RJavaScript/outline.html
-
-- post and update files
--- postFile done
---- skip postFile and updateFile out; this makes no sense
- since everyone knows best (better) how to handle with specific
- files resp. file structures ...
-
-- give more consequent names to cdb list entry's, e.g. $noOfUuids is
- at least a $queryParam
--- done
-
-- localhost is mostly not the right server; maybe it's a good idea to set a
- environment var
--- I think a better idea is to use the R CMD check procedure as a kind of unit
- test (see package notes in package Rd file)
---- bypass all functions by giving no serverName
----- done
-
-- The convention for database naming should be implemented.
--- any API convention should be implemented
-
-- before returning the list the answer should be more evaluated
--- try/catch for all curl calls
-
-
-+++ after this is done I make a R package structure!
-++++ done
-
-- a function cdbDelDoc is missing
--- the function is called cdbDeletDoc
---- done (but not tested well)
-
-- replication?
-
--> mail Duncan:
--><Hi Thomas
--><
-->< Sorry for the long delay in responding to your mail.
--><I have just started teaching again, so there were many
--><things to deal with.
--><
-->< I took a look through the code for your R4CouchDB package.
--><Looks good and reasonably clear and straightforward which is good.
--><
--><Here are some remarks/suggestions. Let me know if they are not clear.
--><
--><cdbIni()
--><
--><Put the dependency on RCurl and RJSONIO in the Description file
--><under Depends: or Imports:
-
-done
-
--><
--><Make the port, server name, db name, etc. arguments
--><with defaults for the cdbIni() function.
-
-ok for port, maybe cdb$serverName <- "localhost", but DBName ...?
-
--><
--><Let the function fail by not being able to load the packages,
--><i.e. use the built-in error mechanism or else use
--><try() and handle the errors by loading the packages.
-todo
--><I.e. don't print the error messages but use stop()
-done
--><with the error message and R's own mechanism will handle
--><it and people know how to deal with that, e.g. options(error = recover),
--><suppress output, etc.
--><
--><cdbListDB() uses getURL().
--><You may want to use getURLContent() for everything, although probably
--><fine.
--><
--><cdbAddDoc()
--><What do you use the reader for in the headerfunction.
--><Doesn't look like you use it. So in that case, getURLContent()
--><will handle the header for you. Even if you do want it, you can get
--><the information about the header from getURLContent()
--><by passing it a dynamic curl reader, e.g.
-->< curl = getCurlHandle()
--><or
-->< curl = cdb$curl
--><
-->< reader = dynCurlReader(curl)
-->< getURLContent(..., header = reader)
--><
-->< reader$header() # to get what was in the header of the HTTP response.
--><
--><You may want to allow the cdb connection object (returned by cdbIni())
--><hava curl object and allow reusing that curl handle each time so as to
--><improve "reconnection" latency, i.e. HTTP 1.1 and keep alive connections.
--><
--><
--><
--><thbo wrote:
--><> > Hi Duncan,
--><> >
--><> > I've introduce unit test using the (really nice) Runit package. This
--><> > leads to some modifications in R4CouchDB.
--><> >
--><> > To have a look on the package you can use
--><> >
--><> > http://github.com/wactbprot/R4CouchDB/
--><> >
--><> > The sources you can download at:
--><> >
--><> > http://wactbprot.github.com/R4CouchDB/
--><> >
--><> > as tar.gz or zip.
--><> >
--><> > It would be nice to hear from you;
--><> >
--><> > Best regards
--><> > Thomas
-
---- apply:
-Since today I can provide a default server for the example sections in the Rd
-files of this package: wactbprot.couchone.com
-This couch is for testing purposes only. Please be careful with traffic
-sensitive data and so on; don't be evil.
-
-Of course the test data will be removed frequently.
-
-to be continued
-
-wactbprot
View
0  example/example.R → demo/example.R
File renamed without changes
Please sign in to comment.
Something went wrong with that request. Please try again.