# Problem with TikZ commutative diagrams: No shape named tikz@f@1-2-2 is known [duplicate]

I am using the Cambridge thesis class file that can be openly found here

In specific I want to use a commutative diagram with tikz-cd but I cannot make it work:

$\begin{tikzcd} A \arrow[rd] \arrow[r] & B \\ \end{tikzcd}$

Is there any way to fix this? The error reads

! Package pgf Error: No shape named tikz@f@1-2-2 is known.

## marked as duplicate by gernot, egreg, Stefan Pinnow, Zarko, CarLaTeXApr 17 '17 at 2:38

• Please post a MWE (Minimal Working Example). – Gregor Perčič Apr 16 '17 at 19:22
• Hi, what is that? I literally cannot make the code above to work (which is very minimal). Thanks – Marion Apr 16 '17 at 19:28
• I already answered. For the next time, the Minimal Working Example is code that compiles but describes the failed accommodations you wanted to make after this step. It makes it way easier for us. But no matter, I answered your question already. – Gregor Perčič Apr 16 '17 at 19:31

The problem is that you targeted the \arrow[rd] into an unexisting cell. I accommodated your snippet of code so it compiles.

\documentclass{article}

\usepackage{mathtools}
\usepackage{tikz-cd}

\begin{document}

$\begin{tikzcd} A \arrow[rd] \arrow[r] & B \\ {} & {} \end{tikzcd}$

\end{document}

Note that I had to make a column for your arrow to go into. I also couldn't signify it with & alone, I had to add two {}'s to signify both cells.

• Oh I see. And I assume the cells are automatically aligned? It works indeed, I completely missed your point. – Marion Apr 16 '17 at 19:32
• No matter. I like to help. I wish you plenty of success with your thesis. – Gregor Perčič Apr 16 '17 at 19:34
• You can use \begin{tikzcd}[nodes in empty cells] for avoiding {} in targeted empty cells. – egreg Apr 16 '17 at 21:06
• Thank you for reminding me. It's been a while since I had to draw some homology commutative diagrams, so, truth be told, I forgot about your function. Though I do think that for a large thesis the regular way is more appropriate as it gives away the "visual" notation of cells immediately. But nevertheless, thank you, Marion's taste in source code is probably different from mine. – Gregor Perčič Apr 16 '17 at 21:20