On Sep 17, 1:00 pm, shinde <[email protected]> wrote: > Thanks jeff... its working now ! :) No problem.
For what its worth, the anonymous variables are legal and have a lifetime of the "full expression", which is the the ending semi-colon (I just checked and verified with one of GNU's stdc++ authors). I would guess that Wei has not experienced the problem. He uses objects such as StringSource a little differently (see the self tests). I haven't been able to get the problem under the debugger, so I don't know what knobs to turn. Jeff > On Sep 17, 3:41 am, Jeffrey Walton <[email protected]> wrote: > > > On Sep 17, 2:33 am, shinde <[email protected]> wrote:> Hello Everyone, > > > > I am currently having trouble setting up crypto++ on ubuntu.. Make > > > works fine, but when i try make install, I get the following error: > > > > # make install > > > ================================================== > > > ERROR : > > > cp *.h /usr/include/cryptopp > > > cp *.a /usr/lib > > > cp *.so /usr/lib > > > cp: cannot stat `*.so': No such file or directory > > > make: *** [install] Error 1 > > > =================================================== > > > The shared object was not built, so it does not exist (and hence, does > > not stat). > > > > But somehow i'm able to verify the installation using the following > > > commands: > > > > > whereis cryptest.exe > > > > whereis libcryptopp.a > > > Again, no shared object (libcryptopp.so). > > > There are some makefile tweaks available > > athttp://sourceforge.net/tracker/?func=detail&aid=3409518&group_id=6152... > > andhttp://sourceforge.net/tracker/?func=detail&aid=3409556&group_id=6152.... > > The tweaks add the static and dynamic objects to the default rule > > (all). They also add a leading hyphen to the CP command so it > > continues on library copy failure (if you build the shared object, but > > not the archive, make will fail before copying the SO during install). > > > > After this I assumed that it is installed and went about trying the > > > samples for RSA given on the wiki.. Even though I had no problem > > > compiling it, I am getting segmentation fault when i try to run it. > > > > //////////////////////////////////////////////// > > > // Generate keys > > > AutoSeededRandomPool rng; > > > > InvertibleRSAFunction params; > > > params.GenerateRandomWithKeySize(rng, 3072); > > > > RSA::PrivateKey privateKey(params); > > > RSA::PublicKey publicKey(params); > > > > string plain="RSA Encryption", cipher, recovered; > > > > //////////////////////////////////////////////// > > > // Encryption > > > RSAES_OAEP_SHA_Encryptor e(publicKey); > > > > StringSource(plain, true, > > > new PK_EncryptorFilter(rng, e, > > > new StringSink(cipher) > > > ) // PK_EncryptorFilter > > > ); // StringSource > > > Try: > > > StringSource sse(plain, true, > > new PK_EncryptorFilter(rng, e, > > new StringSink(cipher) > > ) // PK_EncryptorFilter > > ); // StringSource > > > > //////////////////////////////////////////////// > > > // Decryption > > > RSAES_OAEP_SHA_Decryptor d(privateKey); > > > > StringSource(cipher, true, > > > new PK_DecryptorFilter(rng, d, > > > new StringSink(recovered) > > > ) // PK_DecryptorFilter > > > ); // StringSource > > > Try: > > > StringSource ssd(cipher, true, > > new PK_DecryptorFilter(rng, d, > > new StringSink(recovered) > > ) // PK_DecryptorFilter > > ); // StringSource > > > > cout << "Recovered plain text" << endl; > > > > I think I am getting a segmentation error due to StringSource, but I'm > > > not able to figure out why. I'm new to crypto++ and relatively new to c > > > ++ too. > > > I've got to check if that (not using a variable name) is legal C++. I > > thought it was when the sample was written. If legal, it might be that > > not using a variable does something with sequence points, or GCC is > > optimizing too aggressively. -- You received this message because you are subscribed to the "Crypto++ Users" Google Group. To unsubscribe, send an email to [email protected]. More information about Crypto++ and this group is available at http://www.cryptopp.com.
