Now resolved thanks to a fedora bug reader. The missing information is that although rows are read sequentially top to bottom, groups are read left to right, for some reason within a group the 'bits' are read right to left, left most bit representing the highest code point in that group. Not documented anywhere I could find. My code is now working. Dave On 19 January 2012 13:33, Dave Pawson <dave.pawson@xxxxxxxxx> wrote: > I'm returning to this thread since my understanding of this is somehow wrong. > > On 17 January 2012 14:57, Akira TAGOH <akira@xxxxxxxxx> wrote: >> I'm not sure if I understand a question properly though, U+E059 should >> appears on the output of fc-query when available: >> >> 00e0: 00000000 00000000 02000000 .... >> >> details: >> 0xe059 >> 8 = 0xe0 >> 0xe059 & 0xff / 32 = 2 (+1 column) >> 0xe059 & 0x1f = 25 (bit) >> >> It's the position for U+E059.I'm returning to this thread since I clearly don't understand the matrix. > > The output from fc-query in question is > 00e0: 00000000 00000000 00000070 00000800 > > I'm translating from the matrix to Unicode code points. > > If I understand it correctly, the starting value of each group in this line is: > > E000 E020 E040 ... > Hence for the group 00000070 > 1 32 > The characters are in the range E040..E5F > > Which 'end' of the group represents the lower value please? > The left hand or right hand end? > Is E040 bit 1 or bit 32? > > Is this documented anywhere please? > > TIA > > > > > -- > Dave Pawson > XSLT XSL-FO FAQ. > Docbook FAQ. > http://www.dpawson.co.uk -- Dave Pawson XSLT XSL-FO FAQ. Docbook FAQ. http://www.dpawson.co.uk _______________________________________________ Fontconfig mailing list Fontconfig@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/fontconfig
