[Date Index] [Thread Index] [Author Index]
Re: [Q] Why Integrate[1/x,x] <> Log[Abs[x]] with Mma 2.2 ?
At 08:47 +0200 97.09.19, Denis Barbier wrote: > Hi all, > > it is well known that Integrate[1/x,x] =3D Log[Abs[x]], but Mma 2.2 return= s > Log[x]. > > How can i correct this ? > > Thanks in advance > > Denis Bonjour Denis, I am afraid that what you call "well-known" could be more accurately described by "practical and common convention suited to most real intervals". There are various common methods to force Mathematica to give you Log[Abs[x]] in response to Integrate[1/x,x], for instance: Unprotect[Integrate]; Integrate[1/x_Symbol,x_Symbol,___Rule]:=3DLog[Abs[x]]; Protect[Integrate]; but I cannot recommend making that because the treatment of the arguments patterns of Integrate is organized in an optimized and quite intricate way. In such a powerful (and knowledge-laden) black box function as Integrate, tinkering with it can make you loose a lot by overriding predefined transformations. I hope nevertheless that this answer will help you in what you are currently doing. By the way do you know we have a French Mathematica User Group ? To join, just send a blank-subject message to Listserv at ext.jussieu.fr with the line: sub GROUMF Your name - Your institution Moderators are Eric Lewin and J-C Culioli. I invite all french speaking users of Mathematica to join in. We will be pleased to welcome you ! (We welcome the others as well, but the discussions take place in french) Olivier Gerard Editor in Chief - special issues `Quadrature=B4 <quadrature at technolink.org> =46rench Mathematics Quarterly