I am getting these errors when either directly generating a pdf in LyX or running pdflatex on the generated .tex.
Overfull \hbox (36.72682pt too wide) in paragraph at lines 2163--2166
[][]
[49]
Underfull \vbox (badness 3815) has occurred while \output is active [50 <./obr/
Businger_phim.png (PNG copy)> <./obr/Businger_phih.png (PNG copy)>] [51 <./obr/
Arya11.5.png>]
! Package inputenc Error: Unicode character ŀ (U+0140)
(inputenc) not set up for use with LaTeX.
See the inputenc package documentation for explanation.
Type H <return> for immediate help.
...
l.2288 \chapter
{Konvektivní mezní vrstva}
? [52 <./obr/Garrat3.5.png (PNG copy)>]
! Package inputenc Error: Unicode character ŀ (U+0140)
(inputenc) not set up for use with LaTeX.
See the inputenc package documentation for explanation.
Type H <return> for immediate help.
...
l.2288 \chapter
{Konvektivní mezní vrstva}
?
[53 <./obr/HoltslagNieuwstadt1986fig2.png (PNG copy)>]
! Package inputenc Error: Unicode character ŀ (U+0140)
(inputenc) not set up for use with LaTeX.
See the inputenc package documentation for explanation.
Type H <return> for immediate help.
...
l.2288 \chapter
{Konvektivní mezní vrstva}
?
[54 <./obr/HoltslagNieuwstadt1986fig1.png (PNG copy)>]
Kapitola 11
The file uses the UTF-8 encoding and is written in Czech and uses
\documentclass[oneside,czech]{amsbook}
\usepackage[T1]{fontenc}
\usepackage[utf8]{inputenc}
\usepackage{babel}
\usepackage{units}
\usepackage{textcomp}
\usepackage{mathtools}
\usepackage{url}
\usepackage{enumitem}
\usepackage{amsbsy}
\usepackage{amsthm}
\usepackage{graphicx}
\usepackage[unicode=true]
{hyperref}
As I am aware, the character ŀ does not appear anywhere in the .tex file. I also tried Okteta to search for 0x140 or for 0xC5 0x80 but the search shows nothing. How can I find the origin for this?
The generated pdf looks reasonable, but the table of content for pdf navigation ends before the chapter that is mentioned in the error message (Konvektivní mezní vrstva). I tried changing the name but it does not change anything.
With \errorcontextlines=1000, as suggested by David Carlisle, the error looks like this
! Package inputenc Error: Unicode character ŀ (U+0140)
(inputenc) not set up for use with LaTeX.
See the inputenc package documentation for explanation.
Type H <return> for immediate help.
...
\GenericError ...
\endgroup
\UTFviii@defined ...ax \UTFviii@undefined@err {#1}
\else \PackageError {input...
\@tempa ->{\protect \@upprep ŀ
KÁLOVÁNÍ MIMO PřÍZEMNÍ VRSTVU}
\sectionrunhead ...percasenonmath \@tempa \@tempa
\endgroup \fi
\rightmark ->\expandafter \@secondoftwo \botmark
{}{}
\@oddhead ...\protect \thanks@warning }\rightmark
{}{}\hfil \llap {\thepage }
\@outputpage ...lor \hb@xt@ \textwidth {\@thehead
}\color@endbox }\dp \@temp...
\@opcol ...lumn \@outputdblcol \else \@outputpage
\fi \global \@mparbottom \...
<output> ...specialoutput \else \@makecol \@opcol
\@startcolumn \@whilesw \i...
\newpage ...prevdepth \fi \fi \vfil \penalty -\@M
\clearpage ...opskip \hbox {}\fi \fi \fi \newpage
\write \m@ne {}\vbox {}\pe...
\cleardoublepage ->\clearpage
{\pagestyle {empty}\cleardouble@page }
\chapter ->\if@openright \cleardoublepage
\else \clearpage \fi \thispagestyl...
l.2290 \chapter
{Konvektivní mezní vrstva}
?
this appears to be in the name of a section that is displayed on the top of the page and that contains a capital letter Š (U+0160).
Podobně v modelech můžeme určit toky u zemského povrchu z hodnot teploty
a rychlosti v nejnižší modelové hladině.
\section{Škálování mimo přízemní vrstvu}
Moninova-Obuchovova teorie podobnosti platí v přízemní vrstvě a jen
v určitém rozmezí stability mimo extrémy jak ve stabilním, tak v instabilním
The letter Š is missing in the resulting pdf and FIX-ME was there instead when I redefined the U+0140 character.
Please also note the small ř instead of the capital Ř.
I now noticed, that another letter gets corrupted in a similar location:
from
\section{Empirické podobnostní funkce a jejich důsledky}
That means that Ů/ů U+016E/F was changed into ŏ U+014F.
The texts in the titles within the text itself are correct.
Please also note that when opening the .tex file with UTF-8 encoding, kwrite complains that it contains illegal character and cannot be saved. However, the suspicious section titles appear to be OK.
I now believe that the problem with the pdf navigation is a separate one and I will perhaps ask a separate question.
\errorcontextlines=1000then show the full error message from!to?in a code block so lines are preserved\DeclareUnicodeCharacter{0140}{FIX-ME!!!}and show the resulting output\chapter. What do you have there? Something that produces a few images with a loop, I believe. TeX shows where it's at and that error message means that\chapterhasn't yet been processed.percasenonmaththe class is applying primitive\uppercasebreaking the utf-8 encoding. The details of how to fix depend on your latex version as\MakeUppercasewas reimplemented recently to better work wth Unicode. Do you haveLaTeX2e <2022-11-01> patch level 1?