I modified your pdfs and they are working now : https://www.dropbox.com/sh/9k2x1rz146f0llg/8r4PUF2sGM
I have been facing this problem too and found one solution : I resize the pdfs (say from 100% to 101% or 99%) and they read fine. I do this in foxit advanced pdf editor.
Pdfs converted by amateur softwares cause this problem. The pdf files submitted by you are made by "wkhtmltopdf". Calibre converted pdfs have similar errors. I am now going to learn bits of css to convert my pdfs in a better way.
One more thing - the text in the pdfs is clipped from many pages. (Look at page 24 or 498 of "10601-apprenez-a-programmer-en-java" and compare it with the one i gave you; there were other pages too and they were in all the 4 files). This text will be unclipped automatically by foxit pdf editor. But it is unclipped only till the boundary of page i.e. some text could be missing ..if you wish to recover that text there is one more lengthy procedure.
P.S. The files converted by these softwares may show errors with dictionary as in selecting a whole line instead of a word etc. I wish to convert my epubs into pdf and it seems i will need to learn css properly to do so